Keys to the avian Haemoproteus parasites (Haemosporida, Haemoproteidae)

Gediminas Valkiūnas; Tatjana A Iezhova

doi:10.1186/s12936-022-04235-1

. 2022 Sep 19;21:269. doi: 10.1186/s12936-022-04235-1

Keys to the avian Haemoproteus parasites (Haemosporida, Haemoproteidae)

Gediminas Valkiūnas ^1,^✉, Tatjana A Iezhova ¹

PMCID: PMC9487097 PMID: 36123731

Abstract

Background

Haemoproteus is a sister genus to malaria parasites (Plasmodium), which both belong to the order Haemosporida (Apicomplexa). Parasites of both genera are flourishing in birds, however, Haemoproteus species are noticeably less investigated. This is unfortunate because knowledge about close relatives of malaria pathogens is important for better understanding the evolutionary origin and basic biological features of the entire group of haemosporidian infections. Moreover, recent findings show that Haemoproteus species can cause severe damage of various bird organs due to megalomeronts and other exo-erythrocytic stages. These haemosporidians are remarkably diverse, but remain neglected partly due to difficulties in species identification. Hundreds of Haemoproteus genetic lineages have been reported in birds, and numerous new lineages are found each year, but most remain unidentified to the species level. Numerous new Haemoproteus pathogens were described during the past 20 years. However, keys for their identification are absent. Identification of Haemoproteus species remains a difficult task and is an obstacle for better understanding of the distribution and epidemiology of these parasites. This study aimed to develop comprehensive keys for the identification of described avian Haemoproteus species using morphological features of their blood stages (gametocytes).

Methods

Type and voucher preparations of avian Haemoproteus species were accessed in museums in Europe, Australia and the USA. Gametocytes of most described species were examined, and these data formed a background for this study. The data also were considered from published articles containing parasite species descriptions. The method of dichotomous keys was applied. The most difficult steps in the keys were accompanied with references to the corresponding parasite pictures.

Results

In all, 201 published articles were included in this review. Morphological diagnostic features of gametocytes of all described Haemoproteus species were analysed and compared. Illustrated keys for identification of these parasite species were developed. Available information about the molecular characterization of Haemoproteus parasites was provided.

Conclusion

This review shows that 177 described species of avian Haemoproteus can be distinguished and identified in blood films using morphological characters of their gametocytes and host cells. These species were incorporated in the keys. Information about possible morphologically cryptic parasites was provided. Molecular markers are available for only 42% of the described Haemoproteus parasites, calling for researchers to fill this gap.

Keywords: Key to species, Birds, Molecular characterization, Haemoproteus taxonomy, Plasmodium

Background

Order Haemosporida (Alveolata, Apicomplexa, Sporozoa) unites a diverse group of closely related obligate heteroxenous protists, whose currently are classified in four families—Plasmodiidae, Haemoproteidae, Leucocytozoidae and Garniidae [1–5]. These pathogens have many similar basic features in their life cycles. Mainly, they are transmitted exclusively by dipteran blood-sucking insect (Diptera), in which the sporozoites (invasive stage for vertebrates) develop. Sporozoites initiate the exo-erythrocytic development (exo-erythrocytic merogony or schizogony) in various tissues of vertebrate hosts. Exoerythrocytic merozoites are capable of infecting blood cells, in which gametocytes (the invasive stage for vectors) finally inhabit. Asexual dividing stages (erythrocytic meronts or schizonts) do not develop in Haemoproteus species. Gametocytes are characterized by sexually dimorphic characters. Development in vectors is similar in all haemosporidians; it consists of exflagellation, which is followed by the sexual process of the oogamy, development of motile ookinetes, sporogony, and finally the formation of sporozoites, which complete their maturation in the salivary glands of vectors. Infection of vertebrate hosts occurs actively during the blood meal of vectors, who inject sporozoites into the blood stream. Similarities in life cycles and other features of the biology (particularly ultrastructure) of haemosporidians belonging to different families [3, 5–7] have been supported by molecular phylogenies, which indicate that these parasites are relatives and likely have a common origin [8–12]. In other word, accumulation of new knowledge about different groups of haemosporidians, which are remarkably diverse and flourishing in wildlife [5, 6, 13–15], would be beneficial for better understanding the biology of the entire group of Haemosporida, including agents of diseases of domestic animals and humans, as well as malaria [10].

Molecular phylogenetic data show that Haemoproteus parasites (Haemoproteidae) are a sister group to malaria agents of the genus Plasmodium (Plasmodiidae) [8, 9, 11]. Haemosporidians of both these genera do not digest haemoglobin completely, resulting in the accumulation of residual pigment (haemozoin) in their blood stages [1–3]. This feature unites species of Haemoproteus and Plasmodium and distinguishes them from species of the Leucocytozoidae and Garniidae, which do not produce residual pigment when developing in red blood cells. The life cycles of Haemoproteus and Plasmodium parasites differ mainly due to the inability of the haemoproteids (i) to multiply in blood cells (erythrocytic merogony is absent) and (ii) to complete sporogony in mosquitoes [7]. Haemoproteus parasites are transmitted mainly by Culicoides biting midges (Ceratopogonidae), and a few species are vectored by louse flies (Hippoboscidae) [2, 7, 16]. In spite of these differences, the knowledge on Haemoproteus spp. is important for better understanding evolutionary biology of haemosporidians [17–19], including human malaria parasites of genus Plasmodium [10]. For example, the application of Haemoproteus tartakovskyi genomic information in phylogenetic studies contributes to understanding the evolutionary relationships of Laverania parasites [8].

Haemosporidians of the genus Haemoproteus parasitize only birds and reptiles [1, 3, 6]. These protists remain a neglected group of blood pathogens mainly because they have been traditionally considered to be relatively benign to their hosts [20]. This seems to be true in regard to the blood pathology during haemoproteosis due to the absence of multiplication in blood cells and predominantly light or moderate parasitaemia, which rarely reaches 5%, but usually is less than 1% in wild-caught animals [3]. However, recent studies show that haemoproteids are pathogenic to blood-sucking insects, including mosquitoes, who often die within 12 h after taking heavily infected blood meals due to damage caused by the migration of ookinetes [21–23]. Furthermore, the application of molecular diagnostic tools has proven that large-sized megalomeronts (up to 300 µm and even bigger) develop in many Haemoproteus infections, resulting in the damage of various organs [2, 24–30]. These findings call for research aimed at better understanding the biology of haemoproteids in regard of animal health.

Haemoproteus species are cosmopolitan and often prevalent in birds [3, 5, 14, 31–34]. Molecular studies have revealed over 4600 unique cytochrome b gene lineages of these parasites, and many more likely exist [13] (see MalAvi database http://130.235.244.92/Malavi, accessed in April 2022). Over 170 species of Haemoproteus have been described, including 49 new species described during the past 20 years. Difficulties in species identification using morphological data preclude comprehensive parasite diversity research. This is unfortunate because morphological and molecular data complement each other and are essential in obtaining a true understanding of pathogen diversity in certain bird populations [35, 36], particularly during co-infections of parasites belonging to same genus [37–39]. Species-specific molecular markers are absent for the majority of avian Haemoproteus, and currently they are difficult to design and use due to the vast genetic diversity of these pathogens, most of which remain insufficiently investigated or even non-described in wildlife. Morphological identification using microscopic examination of blood films supplements the information providing by polymerase chain reaction (PCR)-based diagnostic tools and remains important in the research of wildlife haemosporidians [36, 38, 40]. Thus, the available keys for the identification of avian Haemoproteus species [3] should be reworked and supplemented.

This review aimed to simplify the identification of avian Haemoproteus species by developing easy-to-use keys based on morphological characters of gametocytes, the parasite development stage easily accessed due to presence in the peripheral circulation. This article provides comprehensive keys, which include new Haemoproteus species that are not already found in formerly published keys [3]. This should assist academic and veterinary medicine researchers in the identification of Haemoproteus pathogens. The available information about molecular markers (molecular barcodes), which can be used for the detection of described Haemoproteus species and the comparative research, was summarized as well. This study generalizes the over 40-years of experience of the authors in taxonomy of Haemoproteus species, and is designed for researchers who are interested in wildlife pathogens.

Methods

Full-length papers with descriptions of new Haemoproteus species as well as articles with re-descriptions and molecular characterizations of these parasites published in peer-reviewed journals were considered. In all, 201 articles and books were reviewed, and 191 publications containing the representative morphological and/or PCR-based information related to identification of these parasites were cited and incorporated in the References.

Type and voucher preparation as well as gametocyte images of avian Haemoproteus parasites were obtained from the collections of Nature Research Centre (Vilnius, Lithuania), International Reference Centre for Avian Haematozoa (Queensland Museum, Queensland, Australia), the US National Parasite Collection (National Museum of Natural History, Washington DC, USA), Natural History Museum (London, UK), Muséum National d’Histoire Naturelle (Paris, France), Grupo de Estudio Relación Parásito Hospedero, Universidad Nacional de Colombia (Bogotá, Colombia) and individual researchers. All accessed preparations were examined. An Olympus BX61 light microscope (Olympus, Tokyo, Japan) equipped with an Olympus DP70 digital camera and imaging software AnalySIS FIVE (Olympus Soft Imaging Solution GmbH, Münster, Germany) was used to examine preparations and prepare illustrations. It is important to note that the staining quality of some old type specimens of Haemoproteus species, which were obtained from museums, was of insufficient quality mainly due to fading, resulting in some poorly visible diagnostic characters. However, images of the parasites from such preparations were included in this review and were shown in the corresponding species figures if they provided valuable information about the general shape of gametocytes, the size of pigment granules, the mode of influence on host cells and others readily visible taxonomic characters. Use of this information helps to distinguish some species, and also is important for future taxonomic studies being the source of illustrations from valuable type specimens. The quality of these slides could hardly be improved in the future, so worth documentation at this stage. Black and white drawings were also provided to illustrate morphological details of the parasites, of whose high-quality photographs were not available.

Classical dichotomous keys were developed for the identification of Haemoproteus species [3]. Each key consists of steps divided into two alternatives, which identifies the species of a specimen following a series of simple choices that lead the user to the correct name of a given species. The most difficult choices, which might lead to ambiguity, were accompanied with references to the corresponding pictures that further illustrate meaning of the text information. This simplifies the comparison of diagnostic features used in the keys and minimizes possible misunderstanding. All species names in the keys were accompanied with references to the original parasite descriptions, re-descriptions and (or) other publications, which contain description and (or) illustrations of corresponding parasites. Published articles containing valuable morphological descriptions were collected, analysed and cited. These references help to access parasite descriptions and confirm a parasite identification. Information about the barcoding DNA sequences, which can be used for molecular detection and identification of corresponding parasites are also provided.

The experimental observations showed that Haemoproteus species vary in vertebrate host specificity, but the same parasite usually cannot complete life cycle and produce invasive stages (gametocytes) in birds belonging to different orders (see review in [3, p. 69]. Molecular sequence information is in accordance with these empirical data and indicates only rare cases when the same Haemoproteus lineages could be found in birds belonging to different orders [13]. Importantly, the rare reports of the same Haemoproteus lineages in birds of different orders have never been supported by the observation of the corresponding species gametocytes, an invasive parasite stage for vectors, indicating abortive (incomplete) development, which is a dead-end of infection [7]. Thus, morphologically similar parasites in birds belonging to different orders are different species in most cases. This conclusion was confirmed by molecular data [7, 9, 11, 13]. Due to vertebrate host specificity and the resulting restriction of parasite distributions by bird orders, the natural host range of haemosporidians remains helpful in species identification. This provided an opportunity to design separate keys for identification of parasites inhabiting birds of different orders. This approach simplifies parasite identification by minimizing the number of species, which are needed for comparison before making the final conclusion about a species identity. This approach was used in the keys.

Approximately 50% of all described Haemoproteus species parasitize birds of the order Passeriformes. Due to the marked species diversity of haemoproteids in passerines, the keys for their identification were developed for closely related passeriform bird families, which were grouped in suborders or superfamilies as suggested by the current bird phylogenies [41, 42]. Such keys provide the opportunity to easily identify the majority of described parasites of passeriform birds. However, it should be noted that some Haemoproteus parasites of passerines might infect and produce gametocytes in birds belonging to different families within the same order [43], supporting information that host taxonomic characteristic cannot be considered as the main parasite taxonomic character [3, 44]. This might lead to circumstances when a sample under identification could be not found in a certain key. This also certainly will happen if a researcher is dealing with a new (non-described) parasite species. Further comparison of such samples with parasites of most closely related avian groups is suggested before making a final conclusion about the identity of a sample. This is a relatively weak point of the keys for the identification of Haemoproteus species parasitizing passeriform birds. However, experience shows [3] that creating one big key for all parasites of Passeriformes birds would be even more difficult to use due to the need of too many minor morphological characters, which often are difficult to estimate in practical work, particularly during low parasitaemia. This usually make the identification even more complicated in comparison to the approach, which was used here.

Genus Haemoproteus includes two subgenera—Haemoproteus and Parahaemoproteus. Species of these subgenera differ in patterns of sporogony, which occur in louse flies (family Hippoboscidae) and Culicoides biting midges (family Ceratopogonidae), respectively [3, 7]. Haemoproteus and Parahaemoproteus parasites are indistinguishable at the gametocyte stages, but might be present in birds of the same orders. They also sometimes occur in co-infection [3]. The current examples are the parasites of Columbiformes and Suliformes birds [45, 46]. To facilitate identification of species of both subgenera using morphological characters of gametocytes, these parasites were given in the same keys. Most of avian haemoproteid species belong to Parahaemoproteus. All species of subgenus Haemoproteus were indicated in notes to the corresponding keys.

Results

Birds are hosts of various intracellular blood parasites belonging to Apicomplexa. For example, they are infected by haemosporidians of the genera Plasmodium, Leucocytozoon, Garnia, Fallisia and related species of Babesia, Isospora, Lankesterella and Hepatozoon [2–4, 47, 48]. These organisms (Fig. 1a–o) often occur in co-infections. Haemoproteus species can be readily distinguished from all other avian intracellular haematozoa, except Plasmodium parasites, because their blood stages (gametocytes) develop exclusively in red blood cells and always contain refractive pigment granules (haemozoin granules) (Fig. 1a–c), which are absent in all other blood parasites (Fig. 1h–o). Even the smallest Haemoproteus gametocytes contain pigment granules and can be distinguished from other intracellular blood parasites (Fig. 1a), except for Plasmodium species, which sometimes look similar at the gametocyte or early trophozoite stages (Fig. 1d, e). Furthermore, it is important to note that gametocytes of some species of Haemoproteus are similar to the elongate gametocytes of malaria parasites (Fig. 1b–d). Species of Plasmodium can be readily distinguished because they multiply in blood cells and produce meronts (= schizonts), in which invasive merozoites develop (Fig. 1f, g). This is not the case in species of Haemoproteus. Extensive examination of blood films at magnification 500× usually provides opportunity to visualize meronts of Plasmodium even during low parasitaemia and thus to identify malaria infections. Examination of blood films using an oil immersion objective 50× is particularly convenient for this diagnostic procedure. However, an obstacle still might be a low Plasmodium sp. parasitaemia (< 0.001% of infected red blood cells), which however is the obstacle for identification of any haematozoan infection using microscopic examination of blood films and sometimes even sensitive PCR-based methods [38].

Fig. 1 — Main morphological features of blood stages, which are helpful to distinguish *Haemoproteus* parasites (a–c) from other avian intracellular protists (d–o). Young (a) and fully grown (b, c) gametocytes of *Haemoproteus* species. Fully grown gametocytes (**d, e**) and mature erythrocytic meronts (f, g) of *Plasmodium* species. Growing meronts of *Garnia* (h) and *Fallisia* (i) species. Gametocytes (j, k) in roundish host cell (j) and fusiform host cell (k) of *Leucocytozoon* species. Growing meront (l) of *Babesia* species. Merozoite (m) of *Isospora* species. Sporozoite (n) of *Lankesterella* and gamont (o) of *Hepatozoon* species. Note presence of malarial pigment (haemozoin) in species of *Haemoproteus* (a–c) and *Plasmodium* (d–g) and its absence in species of other avian blood parasites (h–o). Elongate gametocytes of malaria parasites belonging to the subgenera *Giovannolaia* and *Huffia* (d) are similar to gametocytes of *Haemoproteus* species in forms, but the gametocytes of malaria parasites usually are more irregular in shape (d) and the outline of their macrogametocyte nuclei often is not so well indistinct (compare b and d). Presence of merogony in blood cells (f, g) clearly shows malaria infection. Long simple arrows—host cell nuclei. Short simple arrows—parasite nuclei. Simple arrowheads—pigment granules. Simple small arrowheads—merozoites. Simple wide short arrows—vacuoles. Triangle arrows—remnants of host cell cytoplasm. Other explanations are given in the text

It should be noted that a shape of fully grown gametocytes is an important character for many Haemoproteus species identifications [2, 3, 49]. Several major gametocyte forms are readily distinguishable and were used in the keys (Fig. 2a–r). These forms are microhalteridial (fully grown gametocytes are small and do not reach the poles of infected erythrocytes, Fig. 2b–f), halteridial (fully grown gametocytes reach and occupy the poles of infected erythrocytes, Fig. 2g–j), circumnuclear (fully grown gametocytes encircle the nuclei of infected erythrocytes completely, Fig. 2o, p) and rhabdosomal (fully grown gametocytes push the nuclei of erythrocytes to polar position and finally enucleate the host cells, Fig. 2q, r). A non-infected erythrocyte was shown in Fig. 2a for comparison with the infected erythrocytes. Sometimes subdivisions of these main forms were also helpful during species identification, and they were mentioned in the keys. These forms are broadly-halteridial (fully grown gametocytes occupy the poles of infected erythrocytes and markedly displace the nuclei laterally, Fig. 2k, l) and close to circumnuclear (fully grown gametocytes encircle the nuclei of infected erythrocytes nearly completely, Fig. 2m, n). It is important to remind in this regard that the form of only fully grown gametocytes was considered in the keys.

Fig. 2 — Main morphological forms of fully grown gametocytes, which are used in keys for identification of *Haemoproteus* species: roundish (b), microhalteridial (c–f), halteridial (g–j) and broadly-halteridial (k, l), close to circumnuclear (m, n), circumnuclear (o, p) and rhabdosomal (q, r). Uninfected erythrocyte (a) was shown for comparison purpose. All pictures show only fully grown gametocytes. Note that form of growing gametocytes often is different from the form of the fully grown gametocytes. That is why the form of young growing gametocytes usually was not mentioned in the keys, except for some rare cases, which were specified in the tables for species identification. The gametocyte forms depend on mode of parasite growth in red blood cells, which often is species-specific and was used in taxonomy. Various variations of halteridial (c–l) and circumnuclear (m–p) gametocyte forms predominate in avian haemoproteids. Roundish form (a) is particularly rare. Rhabdosomal forms (q, r) develop in several *Haemoproteus* species whose fully grown gametocytes enucleate infected erythrocytes. Dumbbell-like form of gametocytes (e) was often mentioned in the keys. Long simple arrows—host cell nuclei. Short simple arrows—parasite nuclei. Simple arrowheads—pigment granules. Other explanations are given in the text

Blood films should be carefully examined before parasite identification. When the most typical growing and fully-grown gametocytes of Haemoproteus are recognized in samples, the next step is the species identification using the keys. The latter were designed in the form of separate tables, each created for a separate group of avian hosts. To facilitate the use of the keys, a guide to all keys has been provided in Table 1. It lists all the tables for Haemoproteus species identification in relation to bird orders or relative families, as grouped in the keys.

Table 1.

Guide to keys of Haemoproteus species according to their avian hosts

Bird taxa^a	Key for parasite species
Order Accipitriformes	Table 2, Fig. 3
Order Anseriformes	Table 3, Fig. 4
Order Apodiformes	Table 4, Fig. 5
Order Bucerotiformes	Table 5, Fig. 6
Order Caprimulgiformes	Table 6, Fig. 7
Order Cariamiformes	Table 7, Fig. 8
Order Cathartiformes	Table 8, Fig. 9
Order Charadriiformes	Table 9, Fig. 10
Order Ciconiiformes	Table 10, Fig. 11
Order Coliiformes	Table 11, Fig. 12
Order Columbiformes	Table 12, Fig. 13
Order Coraciiformes	Table 13, Fig. 14
Order Cuculiformes	Table 14, Fig. 15
Order Falconiformes	Table 15, Fig. 16
Order Galliformes	Table 16, Fig. 17
Order Gruiformes	Table 17, Fig. 18
Order Musophagiformes	Table 18, Fig. 19
Order Otidiformes	Table 19, Fig. 20
Order Pelecaniformes	Table 20, Fig. 21
Order Piciformes	Table 21, Fig. 22
Order Psittaciformes	Table 22, Fig. 23
Order Pterocliformes	Table 23, Fig. 24
Order Strigiformes	Table 24, Fig. 25
Order Suliformes	Table 25, Fig. 26
Order Passeriformes
Suborder Tyranni	Table 26, Fig. 27
Suboder Passeri
Families Meliphagidae, Oriolidae, Pachycephalidae, Vireonidae	Table 27, Fig. 28
Families Aegithinidae, Artamidae, Malaconotidae, Vangidae	Table 28, Fig. 29
Families Corvidae, Dicruridae, Laniidae, Monarchidae	Table 29, Fig. 30
Families Alaudidae, Cisticolidae, Melanocharitidae, Paridae	Table 30, Fig. 31
Families Acrocephalidae, Hirundinidae	Table 31, Fig. 32
Families Leiothrichidae, Phylloscopidae, Pycnonotidae, Sylviidae and Zosteropidae	Table 32, Fig. 33
Families Mimidae, Muscicapidae, Sittidae, Sturnidae and Turdidae	Table 33, Fig. 34
Families Dicaeidae, Estrildidae, Fringillidae, Motacillidae, Nectariniidae, Passeridae, Ploceidae	Table 34, Fig. 35
Families Emberizidae, Icteridae, Parulidae, Passerellidae and Thraupidae	Table 35, Fig. 36

Parasite species	GenBank accession and lineage code (in parentheses)^a	References^b
H. abdusalomovi	Not available	Not available
H. aegithinae	Not available	Not available
H. africanus	Not available	Not available
H. alaudae	Not available	Not available
H. ammoperdix	Not available	Not available
H. anthi	Not available	Not available
H. antigonis	KX223839 (GRUAME01)^c, KX223842 (GRUAME02), KX223843 ((GRUAME03)	[190]
H. apodus	Not available	Not available
H. archilochus	Not available	Not available
H. asymmetricus	MW492355 (TUPHI01)	[63]
H. attenuatus	DQ630007 (LULU01)^d, DQ451431 (LULU01), AY393807 (ROBIN01), KJ488597 (ROBIN01)	[16, 85, 147, 170]
H. balearicae	Not available	Not available
H. balmorali	DQ630008 (SFC1), DQ060770 (SFC1), JX026912 (SFC1), DQ630014 (hCOLL3)	[147, 148, 172]
H. beckeri	Not available	Not available
H. belopolskyi	DQ630006 (HIICT1), JX026904 (HIICT1), JX026909 (HIICT3), MN025422 (HIICT1), KJ627801 (HIICT1), AF254969 (MW1)	[147–149, 159, 160]
H. bennetti	Not available	Not available
H. bilobata	Not available	Not available
H. borgesi	Not available	Not available
H. brachiatus	MK580170 (LK03)	[100]
H. bubalornis	Not available	Not available
H. bucconis	Not available	Not available
H. bucerotis	Not available	Not available
H. bukaka	KX100323 (CRALOU01)	[143]
H. burhini	Not available	Not available
H. buteonis	Not available	Not available
H. calandrellae	Not available	Not available
H. caprimulgi	Not available	Not available
H. catharti	MF953291 (CATAUR01)^e	[69]
H. centropi	Not available	Not available
H. ciconiae	Not available	Not available
H. circumnuclearis	Not available	Not available
H. clamatori	Not available	Not available
H. coatneyi	KT698210 (ARBRU02), KM211350 (ANSOM01), KF537292 (ARBRU01), KT698210 (ARBRU02), KF537309 (ATPAL02), KF537326 (PIOLI03), KF537285 (TANIG01), KF537283 (TANVAS02), KF537327 (ZOCAP13)^f	[62, 182]
H. columbae	KU131584 (COLI03), KU1311585 (COQUI05), KJ644778 (HAECOL01)), KF537314 (HAECOL01)	[62, 81, 191–193]
H. concavocentralis	GQ396708 (HAWF2)	[91]
H. contortus	Not available	Not available
H. coraciae	KU297278 (CORGAR01)	[94]
H. cornuata	Not available	Not available
H. cracidarum	Not available	Not available
H. crumenium	Not available	Not available
H. cublae	Not available	Not available
H. cuculis	Not available	Not available
H. cyanomitrae	EU810741 (CYAOLI03), FJ404696 (CYAOLI05), FJ404698 (CYAOLI06)	[180]
H. danilewskii	DQ451411 (COCOR01)	[85]
H. dicaeus	Not available	Not available
H. dicruri	Not available	Not available
H. dolniki	Not available	Not available
H. elani	Not available	Not available
H. enucleator	DQ659592 (ALCLEU01)^g	[194]
H. erythrogravidus	KF537315 (ZOCAP01), KF537329 (ZOCAP14)	[182]
H. eurylaimus	Not available	Not available
H. eurystomae	Not available	Not available
H. fallisi	Not available	Not available
H. formicarius	Not available	Not available
H. forresteri	Not available	Not available
H. fringillae	DQ060764 (CCF3)	[43, 172]
H. furnarius	Not available	Not available
H. fuscae	EU810722 (CELEC01)	[93]
H. gallinulae	Not available	Not available
H. gavrilovi	KP462688 (MEAPI02)	[91]
H. globulosus	Not available	Not available
H. goodmani	Not available	Not available
H. greineri	Not available	Not available
H. halcyonis	Not available	Not available
H. handai	Not available	Not available
H. herodiadis	Not available	Not available
H. hirundinis	KJ499183 (DELURB01), MN025423 (DELURB2)	[128, 149, 161]
H. homobelopolskyi	HQ386240 (PLOMEL01), HQ386241 (PLOMEL02)	[177]
H. homogeneae	MK580174 (SYAT16)	[100]
H. homohandai	KY783725 (ARCHL01)	[125]
H. homoleiothrichus	KY623721 (TROERY02)^h	[165]
H. homominutus	MK580175 (CUKI1)	[100]
H. homopalloris	MH513601 (PHSIB2)	[168]
H. homopicae	MK580172 (GAGLA07)	[100]
H. homovelans	GU085195 (PICAN02)	[121]
H. indicator	Not available	Not available
H. iwa	JF833050 (FREMIN01)	[131]
H. janovyi	Not available	Not available
H. jenniae	CREFURO1 (JN827318)	[72]
H. kairullaevi	Not available	Not available
H. killangoi	JN661945 (ZOSMAD01), KT777738 (ZOSLAT07), KX604234 (ZOSLAT10), KX604236 (ZOSXAN03)	[93, 164]
H. krylovi	Not available	Not available
H. lairdi	Not available	Not available
H. lanii	DQ451429 (RSB2) DQ630010 (RB1), MN025425 (RB1), JX026907 (RB1), KR049265 (RB1), KU529942 (RB1), DQ630011 (RBS2), DQ630012 (RBS4)	[23, 85, 147–150]
H. larae	AB604310 (SPMAG12), LC230123 (LARCRA02), LC230122 (NUMPHA01)	[74]
H. leiothrichus	KY623720 (TROERY01)^h	[165]
H. lophortyx	Not available	Not available
H. macropigmentatus	Not available	Not available
H. macrovacuolatus	KJ175078 (DENAUT01), KJ592828 (DENAUT01), KJ499987 (CA1017)	[55]
H. madagascariensis	Not available	Not available
H. magnus	DQ451426 (CCF7)	[85]
H. majoris	AF254977 (PARUS1), AY831755 (WW2), MN219405 (PHYBOR04)ⁱ	[28, 43]
H. mansoni	Not available	Not available
H. manwelli	KP462687 (MEAPI01)	[91]
H. megapodius	Not available	Not available
H. meropis	Not available	Not available
H. micronuclearis	HQ386235 (RBQ11), HQ386236 (VILWE1), HQ386237 (PLONIG01), HQ386238 (PLONIG02), HQ386239 (PLONIG03)	[177]
H. minchini	KU160476 (CORCRI01)	[115]
H. minutus	DQ630013 (TURDUS2), DQ060772 (TURDUS2), JX026900 (TURDUS2), DQ060772 (TURDUS2)	[16, 147, 148, 172]
H. monarchus	Not available	Not available
H. montezi	Not available	Not available
H. motacillae	AF495579 (YWT01), AF495580 (YWT02), DQ368371 (YWT03), DQ368372 (YWT05), KC568475 (YWT06)	[16, 181]
H. multipigmentatus	GU296216 (JH003W), GU296215 (ZEGAL05), GU296224 (JH3008W)	[82, 83]
H. multivolutinus	JX275888 (TURTYM01)	[82]
H. neseri	Not available	Not available
H. nettionis	Not available	Not available
H. nipponensis	Not available	Not available
H. nisi	Not available	Not available
H. noctuae	KP794612 (CIRCUM01)	[130]
H. nucleocentralis	MT724553 (TANDES01)	[184]
H. nucleocondensus	JX026901 (GRW1), MN025424 (GRW1)	[148, 149, 158]
H. nucleofascialis	HQ386243 (MALRUB02), HQ386244 (PLOMEL03)	[177]
H. nucleophilus	Not available	Not available
H. orioli	Not available	Not available
H. orizivorae	Not available	Not available
H. ortalidum	KX171627 (PENOBS01)	[195]
H. otocompsae	Not available	Not available
H. pachycephalus	Not available	Not available
H. pallidulus	AY831752 (SYAT03)	[169]
H. pallidus	DQ630004 (hPFC1), JX026899 (hPFC1), DQ630005 (COLL2), DQ060771 (hSFC3)	[43, 147, 148, 172]
H. palloris	AF254971 (WW1)	[91]
H. palumbis	Not available	Not available
H. parabelopolskyi	AY831751 (SYAT2), AY831762 (SYAT16), AF495575 (SYBOR01), JX026902 (SYBOR01), AY831750 (SYAT01), AY831751 (SYAT02), KJ499184 (SYNIS01)	[100, 128, 148, 159]
H. parahirundinis	MT119966 (HIRUS05)	[161]
H.paramultipigmentatus	FJ462657 (COLPAS03), JN788939 (COLPAS05)	[82]
H. paranucleophilus	HQ386242 (MALRUB01)	[177]
H. paraortalidum	MH036944 (ABUJAC01)	[108]
H. parus	Not available	Not available
H. passeris	DQ451420 (P37), DQ451421 (P102), DQ451422 (P89), DQ451423 (P138), HM146898 (PADOM05), GU065651 (PAHIS01)	[85, 196]
H. pastoris	KJ499185 (LAMPUR01), KU752568 (LAMPUR01)	[128, 173]
H. payevskyi	AF254968 (RW01), DQ630009 (RW01), JX026905 (RW01)	[147, 148]
H. pelouroi	Not available	Not available
H. philippinensis	Not available	Not available
H. picae	Not available^j	Not available
H. pittae	Not available	Not available
H. plataleae	Not available	Not available
H. porzanae	Not available	Not available
H. pratasi	Not available	Not available
H. psittaci	Not available	Not available
H. pteroclis	Not available	Not available
H. ptilotis	KP721986 (LICHRYS01), JX021554 (LICFRE03), KP721990 (MELLEW03), KP721987 (MELALB01), KP721988 (MELALB02), KP721989 (MELALB03), AY714177 (MELLEW01), KP721992 (NMIN01)	[138]
H. pulcher	OL906298 (CARCRI02) ^k	[67]
H. queleae	Not available	Not available
H. quiscalus	Not available	Not available
H. rileyi	Not available	Not available
H. rotator	Not available	Not available
H. sacharovi	JX073258 (MODO1)	[45]
H. sanguinis	DQ451409 (BUL1), DQ451410 (BUL4)	[85]
H. scolopaci	Not available	Not available
H. sequeirae	Not available	Not available
H. sittae	Not available	Not available
H. skuae	Not available	Not available
H. souzalopesi^l	Not available	Not available
H. stableri	Not available	Not available
H. stellaris	Not available	Not available
H. syrnii	KJ451480 (OTSCO05), KF279523 (STAL02), DQ451424 (STSEL01), KP794611 (CULKIB01)^m	[128–130]
H. tartakovskyi	AY393806 (SISKIN01), JX026908 (SISKIN01), JX026903 (HAWF1), GU289671 (ALARV1), GU289672 (ALARV2), GU289673 (ALARV3)	[148, 150, 172, 179]
H. telfordi	Not available	Not available
H. thereicerycis	Not available	Not available
H. timalus	Not available	Not available
H. tinnunculiⁿ	MK580171 (FALSUB01)	[100]
H. trochili	Not available	Not available
H. trogonis	Not available	Not available
H. turtur	DQ451425 (STSEN1)	[85]
H. tyranni	Not available	Not available
H. undulatus	Not available	Not available
H. upupae	Not available	Not available
H. uraeginthus	Not available	Not available
H. vangii	Not available	Not available
H. vacuolatus	EU770153 (ANLAT02)	[167]
H. valkiunasi	GQ404559 (FREAND01)	[46]
H. velans	MH311671 (MELCAR01), MH311672 (MELSTR01)	[120]
H. vireonis	FJ168561 (VIGIL01), KF482350 (VIOLI05), KF537331 (VIOLI06)	[62, 136]
H. wenyoni	Not available	Not available
H. witti	JQ988105 (TROAED20), KF767420 (TROAED20)	[61, 62]
H. xantholaemae	Not available	Not available
H. zosteropis	JX021550 (ZOSLAT04), KX604237 (ZOSXAN02), KX604235 (ZOSLAT11), FJ664153 (ZOSSTE01)	[164]

Step	Features and species
1 (6)	Fully grown gametocytes, which markedly displace nuclei of infected erythrocytes laterally, often close to the erythrocyte envelope, develop and are common (Fig. 3a, b, d)
2 (3)	Both broadly halteridial (Fig. 3a, d) and circumnuclear (Fig. 3c, f) fully grown gametocytes occur simultaneously
	………………………………………………………………………………………………………………….	*H. janovyi* (Fig. 3a–c) [3, 50]
3 (2)	Fully grown gametocytes predominantly are broadly halteridial (Fig. 3a, d). Circumnuclear fully grown gametocytes (Fig. 3c, f) are absent or occur only occasionally
4 (5)	Fully grown gametocytes fill the infected erythrocytes up to their poles (Fig. 3d)
	………………………………………………………………………………………………………………….	*H. elani* (Fig. 3d) [3, 51, 52]
5 (4)	Fully grown gametocytes predominantly do not fill the infected erythrocytes up to their poles (Fig. 3e)
	………………………………………………………………………………………………………………….	*H. buteonis* (Fig. 3e) [3, 53]
6 (1)	Fully grown gametocytes, which markedly displace nuclei of erythrocytes laterally, often close to the erythrocyte envelope (Fig. 3a, b, d), are absent or occur only occasionally. The majority of gametocytes grow around nuclei of the infected erythrocytes (Fig. 3f) and usually do not markedly displace the nuclei laterally (NDR is close to 1). Circumnuclear gametocytes (Fig. 3c, f) predominate among the fully grown gametocytes
	………………………………………………………………………………………………………………….	*H. nisi* (Fig. 3f) [3, 53, 54]

Step	Features and species
1 (2)	Macrogametocytes often contain one or several large (2.5 μm and bigger) vacuoles (Fig. 4a, b)
	………………………………………………………………………………………………………………….	*H. macrovacuolatus* (Fig. 4a, b) [55]
2 (1)	Macrogametocytes do not contain large (2.5 μm and bigger) vacuoles (Fig. 4a, b)
3 (4)	Fully grown gametocytes predominantly grow around nuclei of infected erythrocytes and do not displace or slightly displace (Fig. 4c, d) the nuclei laterally; they markedly enclose the nuclei with their ends and can completely encircle the nuclei (Fig. 4c). Gametocytes markedly displacing erythrocyte nuclei laterally might occur occasionally during high parasitemia, but never predominate
	………………………………………………………………………………………………………………….	*H. greineri* (Fig. 4c, d) [3, 56]
4 (3)	Fully grown gametocytes predominantly markedly displace the nuclei of infected erythrocytes laterally (Fig. 4e, f); they usually slightly enclose the nuclei with their ends but usually do not encircle the nuclei completely. Circumnuclear fully grown gametocytes might occur occasionally during high parasitemia, but never predominate
	………………………………………………………………………………………………………………….	*H. nettionis* (Fig. 4e, f) [3, 57, 58]

Step	Features and species
1 (2)	Advanced gametocytes grow around nuclei of infected erythrocytes and markedly enclose the nuclei with their ends; fully grown gametocytes can completely encircle the nuclei and occupy all available cytoplasmic space in the erythrocytes (Fig. 5a, b)
	………………………………………………………………………………………………………………….	*H. archilochus* (Fig. 5a, b) [3, 59]
2 (1)	Advanced gametocytes grow along the nuclei of infected erythrocytes, but only slightly enclose the nuclei with their ends (Fig. 5c-h). Fully grown gametocytes do not completely encircle the nuclei of infected erythrocytes
3 (4)	The average number of pigment granules in fully grown gametocytes is greater than 18
	………………………………………………………………………………………………………………….	*H. witti* (Fig. 5c, d) [3, 60–63]
4 (3)	The average number of pigment granules in fully grown gametocytes is less than 18
5 (6)	Growing gametocytes of dumbbell-like shape (Fig. 5e, f) are present
	………………………………………………………………………………………………………………….	*H. apodus* (Fig. 5e, f) [3, 64]
6 (5)	Growing gametocytes of dumbbell-like shape (Fig. 5e, f) are absent
	………………………………………………………………………………………………………………….	*H. trochili* Fig. (Fig. 5g, h) [3, 60]

Step	Features and species
1 (2)	Nuclei assume predominantly central or subcentral position in the fully grown macrogametocytes (Fig. 6a); subterminal position of nuclei is not characteristic. Pigment granules tend to aggregate in conspicuous groups (Fig. 6a)
	………………………………………………………………………………………………………………….	*H. upupae* (Fig. 6a) [3, 51] ^a
2 (1)	Nuclei assume predominantly subterminal position in the fully grown macrogametocytes (Fig. 6b); central or subcentral position of nuclei is not characteristic. Pigment granules are randomly scattered (Fig. 6b); they do not tend to aggregate in conspicuous groups
	………………………………………………………………………………………………………………….	*H. bucerotis* (Fig. 6b) [3, 65]

Step	Features and species
1	One species has been described. The parasite with pleomorphic gametocytes (Fig. 7a–c). Fully grown gametocytes usually slightly enclose the nuclei of infected erythrocytes with their ends and markedly displace the nuclei laterally, but do not encircle them completely (Fig. 7b). However, the gametocytes sometimes also completely encircle the nuclei of erythrocytes and occupy all available cytoplasmic space in the host cells (Fig. 7c). Advanced growing gametocytes often do not adhere to nuclei of erythrocytes (Fig. 7a). The average number of pigment granules is about 20 in fully grown gametocytes
	………………………………………………………………………………………………………………….	*H. caprimulgi* (Fig. 7a–c) [3, 66]

Step	Features and species
1	One species has been described. Advanced growing gametocytes are closely appressed to the envelop of erythrocytes, but often do not touch nuclei of the erythrocytes (Fig. 8b). Form of fully grown gametocytes is close to circumnuclear (Fig. 8a). Fully grown gametocytes cause rounding of the erythrocyte nuclei, which become roundish or even circular in shape (Fig. 8a, b). The average number of pigment granules in fully grown macrogametocytes is greater than in microgametocytes
	………………………………………………………………………………………………………………….	*H. pulcher* (Fig. 8a, b) [67]

Step	Features and species
1	One species has been described. Gametocytes are thick, halteridial, with entire margins; they displace host cell nuclei laterally. Macrogametocyte nucleus is close to central position. Growing gametocytes usually do not adhere to erythrocyte nuclei. The number of pigment granules is close to 25 in fully grown gametocytes; the pigment granules are of medium size, scattered throughout the cytoplasm (Fig. 9a, b)
	………………………………………………………………………………………………………………….	*H. catharti* (Fig. 9a, b) [68, 69]

Step	Features and species
1 (2)	Fully grown gametocytes, which rotate erythrocyte nuclei 45° to 90° to the normal axis (Fig. 10a–c), are common
	………………………………………………………………………………………………………………….	*H. rotator* (Fig. 10a–c) [3, 70]
2 (1)	Fully grown gametocytes, which rotate erythrocyte nuclei 45° to 90° to the normal axis (Fig. 10a–c), are absent or appear only incidentally
3 (9)	Fully grown gametocytes, which completely encircle the nuclei of erythrocytes, are present (Fig. 10d, e, h). Close to circumnuclear (Fig. 10f, g, i) and completely circumnuclear (Fig. 10d, e, h) forms are common at final stage of gametocyte growth
4 (12)	Fully grown circumnuclear macrogametocytes and microgametocytes are closely appressed to the nuclei of infected erythrocytes and occupy all available cytoplasmic space in the erythrocytes (Fig. 10d, h)
5 (6)	Nucleoli are readily visible in nuclei of advanced and fully grown macrogametocytes (Fig. 10d)
	………………………………………………………………………………………………………………….	*H. scolopaci* (Fig. 10d, e) [3, 71]
6 (5)	Nucleoli (Fig. 10d) are invisible in nuclei advanced and fully grown macrogametocytes
7 (8)	Fully grown gametocytes contain predominantly roundish or slightly oval, of approximately uniform size and shape pigment granules (Fig. 10f, g). Elongate rod-like (thin) pigment granules (Fig. 10h, i) usually are absent
	………………………………………………………………………………………………………………….	*H. jenniae* (Fig. 10f, g) [72]
8 (7)	Fully grown gametocytes contain variable in size and shape pigment granules; roundish, oval and elongate rod-like (thin) pigment granules develop, but oval and rod-like pigment granules readily predominate (Fig. 10h, i)
	………………………………………………………………………………………………………………….	*H. larae* (Fig. 10h, i) [3, 73, 74]
9 (3)	Fully grown gametocytes, which completely encircle the nuclei of erythrocytes (Fig. 10d, e, h), are absent. Advanced and fully grown gametocytes are halteridial (Fig. 10j, k)
10 (11)	Fully grown gametocytes contain large (1.0 to 1.5 µm) pigment granules (Fig. 10 j)
	………………………………………………………………………………………………………………….	*H. abdusalomovi* (Fig. 10j) [3, 75]
11 (10)	Fully grown gametocytes do not contain large (1.0 to 1.5 µm) pigment granules
	………………………………………………………………………………………………………………….	*H. burhini* (Fig. 10k) [3, 65]
12 (4)	Fully grown circumnuclear macrogametocytes are not closely appressed to the nuclei of infected erythrocytes and do not occupy all available cytoplasmic space in the erythrocytes (Fig. 10m, o, p). More or less evident unfilled space is present between fully grown circumnuclear macrogametocytes and nuclei of infected erythrocytes (Fig. 10m, o, p). Shape of fully grown macrogametocytes is markedly irregular (wriggled) (Fig. 10m, o, p)
13 (14)	Cytoplasm of advanced and fully grown gametocytes is markedly vacuolated, with several large (bigger than 1 µm) vacuoles often present (Fig. 10l, n). Fully grown microgametocytes, which adhere to the nuclei of infected erythrocytes (Fig. 10n), are common
	………………………………………………………………………………………………………………….	*H. skuae* (Fig. 10l–n) [76]
14 (13)	Cytoplasm of advanced and fully grown gametocytes is not markedly vacuolated; large (bigger than 1 µm in diameter) vacuoles (Fig. 10l, n) usually are absent, but a few small vacuoles might occur. Fully grown gametocytes, which adhere to the nuclei of infected erythrocytes (Fig. 10n), are absent or occur only occasionally
	………………………………………………………………………………………………………………….	*H. contortus* (Fig. 10o, p) [3, 70]

Step	Features and species
1 (2)	Fully grown gametocytes are broadly halteridial (Fig. 11a). Nuclei of fully grown macrogametocytes are median in position, often ribbon-like in form, closely appressed to the erythrocyte nuclei (Fig. 11a). One or several circular large (about 1 µm and bigger) vacuoles are often present in the macrogametocyte cytoplasm (Fig. 11a)
	………………………………………………………………………………………………………………….	*H. crumenium* (Fig. 11a) [3, 77]
2 (1)	Fully grown gametocytes are halteridial (Fig. 11b). Nuclei of fully grown macrogametocytes are median or submedian in position, markedly variable in form and position, usually do not closely appressed to the erythrocyte nuclei (Fig. 11b). Circular conspicuous (about 1 µm and bigger) vacuoles (Fig. 11a) usually are absent from the macrogametocyte cytoplasm, but small vacuoles (less than 0.5 µm) might be present
	………………………………………………………………………………………………………………….	*H. ciconiae* (Fig. 11b) [78]

Step	Features and species
1	One species has been described. Advanced gametocytes are wavy (undulated) or ameboid in outline (Fig. 12a); they grow around the nuclei of infected erythrocytes and markedly enclose them with their ends, but do not displace or only slightly displace the nuclei laterally (Fig. 12b). Growing gametocytes are usually appressed to the envelope of erythrocytes but do not touch the nuclei of the erythrocytes along their entire margin (Fig. 12a, b). Circumnuclear gametocytes occur
	………………………………………………………………………………………………………………….	*H. undulatus* (Fig. 12a, b) [3, 79]

Step	Features and species
1 (12)	Fully grown gametocytes are predominantly halteridial in form (Fig. 13a–i); they outwardly are not similar to gametocytes of Leucocytozoon sp.
2 (9)	Volutin is prominent and readily visible in fully grown gametocytes (Fig. 13a–f) ^b
3 (8)	Volutin is arranged as discrete granules (volutin granules) (Fig. 13a–e), but does not overfill more or less evenly the entire gametocyte cytoplasm
4 (5)	Large (1 µm and bigger) discrete volutin granules are present in fully grown microgametocytes (Fig. 13b). Pigment granules (hemozoin) are present inside volutin granules; this is particularly well visible in microgametocytes (Fig. 13b)
	………………………………………………………………………………………………………………….	*H. columbae*^c (Fig. 13a, b) [3, 45, 62, 80–83]
5 (4)	Large (1 µm and bigger) volutin granules (Fig. 13b) are absent from fully grown microgametocytes. Pigment granules (hemozoin) are absent from volutin granules (Fig. 13c–e)
6 (7)	Fully grown macrogametocytes and microgametocytes markedly displace nuclei of infected erythrocytes laterally (Fig. 13c, d). NDR is less than 0.5
	………………………………………………………………………………………………………………….	*H. turtur*^d (Fig. 13c, d) [3, 45, 82–85]
7 (6)	Fully grown macrogametocytes and microgametocytes do not displace or only slightly displace nuclei of infected erythrocytes laterally (Fig. 13e). NDR is greater than 0.5
	………………………………………………………………………………………………………………….	*H. palumbis*^c (Fig. 13e) [3, 82, 83, 86]
8 (3)	Volutin (Fig. 13a-e) is not arranged as discrete granules (volutin granules); it is arranged as prominent irregular clamps, which often overlap and usually overfill more or less evenly the entire cytoplasm in fully grown gametocytes (Fig. 13f)
	………………………………………………………………………………………………………………….	*H. multivolutinus*^c (Fig. 13f) [82]
9 (2)	Volutin is absent from fully grown gametocytes or difficult to visualize (Fig. 13g, i) ^b
10 (11)	The average number of pigment granules in fully grown macrogametocytes and microgametocytes is greater than 30. The earlies young gametocytes look like elongate slender bodies (Fig. 13h)
	………………………………………………………………………………………………………………….	*H. multipigmentatus*^c (Fig. 13g, h) [45, 82, 83]
11 (10)	The average number of pigment granules in fully grown macrogametocytes and microgametocytes is less than 30. The earlies young gametocytes look like broadly ovoid bodies (Fig. 13j)
	………………………………………………………………………………………………………………….	*H. paramultipigmentatus*^c (Fig. 13i, j) [82]
12 (1)	Fully grown gametocytes are highly pleomorphic; many of them are outwardly similar to gametocytes of Leucocytozoon (Fig. 13k, l); they cause marked deformation of the infected erythrocytes (Fig. 13k, l). Pigment granules are few (less than 10 on average), tiny (often dust-like in appearance, see Fig. 13l)
	………………………………………………………………………………………………………………….	*H. sacharovi*^d (Fig. 13k, l) [3, 45, 80, 82, 83, 87]

Step	Features and species
1 (6)	Fully grown gametocytes, which displace the nuclei of infected erythrocytes towards one pole of the erythrocytes (Fig. 14a, e) are present; some fully grown gametocytes can enucleate the erythrocytes (Fig. 14b–d)
2 (5)	The average number of pigment granules in gametocytes is greater than 11. Macrogametocytes usually do not possess a conspicuous large (bigger than 1 µm in diameter) vacuole (Fig. 14e). Microgametocytes with a highly ameboid outline (Fig. 14f) are not characteristic
3 (4)	Erythrocytes with fully grown gametocytes are significantly atrophied in width in comparison to uninfected erythrocytes
	………………………………………………………………………………………………………………….	*H. lairdi* (Fig. 14a, b) [3, 88]
4 (3)	Erythrocytes with fully grown gametocytes are not significantly atrophied in width (Fig. 14c) in comparison to uninfected erythrocytes
	………………………………………………………………………………………………………………….	*H. enucleator* (Fig. 14c, d) [3, 89]
5 (2)	The average number of pigment granules in gametocytes is less than 11. Macrogametocytes frequently possess one conspicuous large (greater than 0.5 µm in diameter) vacuole (Fig. 14e). Microgametocytes with a highly ameboid outline (Fig. 14f) are common
	………………………………………………………………………………………………………………….	*H. gavrilovi* (Fig. 14e, f) [3, 90, 91]
6 (1)	Fully grown gametocytes, which displace the nuclei of infected erythrocytes toward one pole of the erythrocytes (Fig. 14a, e) are absent; fully grown gametocytes do not enucleate the erythrocytes
7 (10)	Fully grown gametocytes markedly enclose the nuclei of infected erythrocytes with their ends (Fig. 14g–k) and finally can completely encircle the nuclei (Fig. 14g, i). Circumnuclear gametocytes develop (Fig. 14g, h)
8 (9)	Circumnuclear fully grown gametocytes are common (Fig. 14i). Advanced growing gametocytes usually do not adhere to nuclei of infected erythrocytes (Fig. 14h). The average number of pigment granules in gametocytes is greater than 20
	………………………………………………………………………………………………………………….	*H. fuscae* (Fig. 14g–i) [3, 92, 93]
9 (8)	Fully grown gametocytes markedly enclose the nuclei of infected erythrocytes with their ends (Fig. 14j, k) and can completely encircle the nuclei of erythrocytes, but the circumnuclear fully grown gametocytes are uncommon and might appear only at final stages of gametocyte maturation. Advanced growing gametocytes are closely appressed to nuclei of the infected erythrocytes (Fig. 14j). The average number of pigment granules in gametocytes is less than 20
	………………………………………………………………………………………………………………….	*H. coraciae* (Fig. 14j, k) [3, 90, 94]
10 (7)	Fully grown gametocytes only slightly enclose the nuclei of infected erythrocytes with their ends (Fig. 14l–t) and never encircle them completely. Fully grown gametocytes are halteridial or broadly halteridial (Fig. 14l–t). Circumnuclear gametocytes (Fig. 14g, i) do not develop
11 (16)	Nuclei of microgametocytes are diffuse (non-condensed) (Fig. 14p, r); size of microgametocyte nuclei is greater than in macrogametocytes (compare Fig. 14p, r with Fig. 14q)
12 (13)	The average number of pigment granules in gametocytes is greater than 15
	………………………………………………………………………………………………………………….	*H. eurystomae* (Fig. 14l–n) [3, 95]
13 (12)	The average number of pigment granules in gametocytes is less than 15
14 (15)	Fully grown gametocytes are broadly halteridial (Fig. 14o, p). The cytoplasm of advanced macrogametocytes often contains conspicuous small (usually less than 1 µm in diameter) circular vacuoles (Fig. 14o). The average width of fully grown gametocytes is greater than 4 µm. The average NDR is 0.5 or less
	………………………………………………………………………………………………………………….	*H. manwelli* (Fig. 14o, p) [3, 88, 91]
15 (14)	Fully grown gametocytes are halteridial (Fig. 14q, r). The cytoplasm of advanced macrogametocytes usually do not contains conspicuous small circular vacuoles (Fig. 14o). The average width of fully grown gametocytes is less than 4 µm. The average NDR is greater than 0.5
	………………………………………………………………………………………………………………….	*H. meropis* (Fig. 12q, r) [3, 88]
16 (11)	Nuclei of microgametocytes are small (condensed) (Fig. 14t); size of nuclei in macrogametocytes Fig. 14s) and microgametocytes (Fig. 14t) are similar
	………………………………………………………………………………………………………………….	*H. halcyonis* (Fig. 14s, t) [3, 92]

Step	Features and species
1 (2)	Gametocytes grow around the nuclei of infected erythrocytes (Fig. 16a–c). Advanced growing gametocytes are closely appressed to envelope of the infected erythrocytes (Fig. 16a). Growing gametocytes markedly displace the erythrocyte nuclei laterally (Fig. 16a). Outline of growing gametocytes vary from even to amoeboid. Fully grown gametocytes can completely encircle erythrocyte nuclei (Fig. 16b, c) and occupy all available cytoplasmic space in the erythrocytes, but broadly halteridial fully grown gametocytes also are present. Erythrocytes containing fully-grown circumnuclear gametocytes often assume roundish form (Fig. 16c)
	………………………………………………………………………………………………………………….	*H. tinnunculi* (Fig. 16a–c)[3, 99, 100]
2 (1)	Gametocytes grow around the nuclei of infected erythrocytes (Fig. 16d–f). Advanced growing gametocytes often do not adhere to envelope of erythrocytes (Fig. 16d). Growing gametocytes do not displace or only slightly displace the erythrocyte nuclei laterally (Fig. 16d, e). Growing gametocytes usually are highly irregular (Fig. 16d) or amoeboid in outline, with readily visible prominent and variable in form outgrowths (Fig. 16e). Fully grown gametocytes nearly completely (Fig. 16f) or even completely encircle the nuclei of infected erythrocytes, they usually do not occupy all available cytoplasmic space in the erythrocytes (Fig. 16f). Infected erythrocytes do not assume roundish form (compare Fig. 16c with Fig. 16f)
	………………………………………………………………………………………………………………….	*H. brachiatus* (Fig. 16d–f) [3, 100]

Step	Features and species
1 (16)	Advanced growing and fully grown macrogametocytes do not contain one large (bigger than 1 µm in diameter) circular vacuole (Fig. 17j–l)
2 (7)	Gametocytes grow around nuclei of infected erythrocytes; they markedly enclose the nuclei with their ends (Fig. 17a–d) and finally encircle the nuclei completely. Fully grown gametocytes are circumnuclear (Fig. 17a, b, d)
3 (6)	Fully grown gametocytes usually do not occupy all available cytoplasmic space in infected erythrocytes. A more or less evident unfilled space (a ‘cleft’) usually is present between the fully grown gametocyte and the nucleus of erythrocyte (Fig. 17a–c)
4 (5)	The average number of pigment granules in macrogametocytes is greater than 15
	………………………………………………………………………………………………………………….	*H. lophortyx* (Fig. 17a,b) [3, 102]
5 (4)	The average number of pigment granules in macrogametocytes is less than 15
	………………………………………………………………………………………………………………….	*H. stableri* (Fig. 17c) [3, 103]
6 (3)	Fully grown gametocytes occupy all available cytoplasmic space in infected erythrocytes (Fig. 17d). An unfilled space (a ‘cleft’) (Fig. 17a, b) is absent between the fully grown gametocyte and nucleus of the erythrocyte
	………………………………………………………………………………………………………………….	*H. mansoni* (probable syn. H. meleagridis) (Fig. 17d) [3, 104]
7 (2)	Fully grown gametocytes do not encircle the nuclei of infected erythrocytes completely. Fully grown gametocytes are halteridial (Fig. 17e–i)
8 (14)	The average number of pigment granules in fully grown macrogametocytes is greater than 15
9 (15)	The outline of fully grown gametocytes predominantly is even (Fig. 17e–g)
10 (13)	Growing gametocytes with a highly constricted central portion, causing a ‘dip’ and presenting a dumbbell-like shape (Fig. 2e), are absent
11 (12)	Fully grown macrogametocytes markedly displace the nuclei of infected erythrocytes laterally (Fig. 17e); the average NDR is less than 0.7
	………………………………………………………………………………………………………………….	*H. pratasi* (Fig. 17e) [3, 105]
12 (11)	Fully grown gametocytes do not displace or slightly displace the nuclei of infected erythrocytes laterally (Fig. 17f); the average NDR is greater than 0.7
	………………………………………………………………………………………………………………….	*H. ammoperdix* (Fig. 17f) [3]
13 (10)	Growing gametocytes with a highly constricted central portion, causing a ‘dip’ and presenting a dumbbell-like shape (Fig. 2e), are present
	………………………………………………………………………………………………………………….	*H. megapodius*^a [3, 106]
14 (8)	The average number of pigment granules in fully grown macrogametocytes is less than 15
	………………………………………………………………………………………………………………….	*H. rileyi* (Fig. 17g) [3, 105]
15 (9)	The outline of fully grown and advanced gametocytes predominantly is highly ameboid (Fig. 17h, i)
	………………………………………………………………………………………………………………….	*H. cracidarum* (Fig. 17h, i) [3, 107]
16 (1)	Advanced growing and fully grown macrogametocytes often contain one large (bigger than 1 µm in diameter) circular vacuole (Fig. 17j–l)
17 (18)	Fully grown gametocytes of roundish or various oval forms (Fig. 17k, l) are absent. Vacuole-like unstained spaces might be present on one or both ends of maturing and fully grown macrogametocytes (Fig. 17j). Gametocytes are overfilled with prominent distinct volutin granules (Fig. 17j)
	………………………………………………………………………………………………………………….	*H. paraortalidum* (Fig. 17j) [108]
18 (17)	Fully grown gametocytes of roundish or various oval forms (Fig. 17k, l) are present. Vacuole-like unstained spaces (Fig. 17j) are absent on both ends of growing or fully grown macrogametocytes. Gametocytes contain volutin, which is dispersed (Fig. 17k, l)
	………………………………………………………………………………………………………………….	*H. ortalidum* (Fig. 17k, l) [3, 109]

Step	Features and species
1 (6)	Growing dumbbell-shaped gametocytes (Fig. 18f, g) are absent
2 (3)	Growing and fully grown gametocytes are elongate and slender (Fig. 18a, b); they do not display or only slightly display nuclei of the infected erythrocytes laterally (Fig. 18a, b). The average NDR is close to 1
	………………………………………………………………………………………………………………….	*H. balearicae* (Fig. 18a, b) [3, 110]
3 (2)	Growing and fully grown gametocytes are broadly halteridial (Fig. 18c, d); they do not assume slender form (Fig. 18a, b). Fully grown gametocytes markedly display nuclei of the infected erythrocytes laterally (Fig. 18d, e); the average NDR is less than 0.5
4 (5)	Average number of pigment granules in macrogametocytes is greater than 25
	………………………………………………………………………………………………………………….	*H. gallinulae* (Fig. 18c, d) [3, 111, 112]
5 (4)	Average number of pigment granules in macrogametocytes is less than 25
	………………………………………………………………………………………………………………….	*H. antigonis* (Fig. 18e) [3, 113]
6 (1)	Growing dumbbell-shaped gametocytes (Fig. 18f, g) are common
	………………………………………………………………………………………………………………….	*H. porzanae* (Fig. 18f–h) [3, 112]

Step	Features and species
1 (2)	Fully grown gametocytes are halteridial; they only slightly enclose nuclei of infected erythrocytes with their ends (Fig. 19a, b). Both ends of advanced gametocytes usually are more or less narrowed in comparison to the widths of the gametocytes (Fig. 19a, b). One or both ends of advanced gametocytes often are pointed (Fig. 19a, b)
	………………………………………………………………………………………………………………….	*H. montezi* (Fig. 19a, b) [3, 114]
2 (1)	Fully grown gametocytes are halteridial; they only slightly enclose nuclei of infected erythrocytes with their ends (Fig. 19c). Both ends of advanced gametocytes usually are not narrowed in comparison to the widths of the gametocytes (Fig. 19c). Both ends of advanced gametocytes usually are approximately similarly rounded (Fig. 19c)
	………………………………………………………………………………………………………………….	*H. minchini* (Fig. 19c) [115]

Step	Features and species
1	One species has been described. Growing gametocytes are closely appressed to the erythrocyte envelope but usually do not touch the erythrocyte nuclei, and as a result, a more or less evident unfilled space (a ‘cleft’) is present between the gametocytes and the erythrocyte nuclei (Fig. 20a). Fully grown gametocytes are closely appressed both to the nuclei and envelope of erythrocytes; they completely encircle the nuclei and occupy all available cytoplasmic space in the erythrocytes (Fig. 20b). The average number of pigment granules is about 17 in macrogametocytes
	………………………………………………………………………………………………………………….	*H. telfordi* (Fig. 20a, b) [3, 113]

Step	Features and species
1 (2)	Fully grown gametocytes occupy all or nearly all cytoplasmic space in infected erythrocytes, including poles of the erythrocytes (Fig. 21a, b); they vary from broadly halteridial (Fig. 21a, c) to circumnuclear (Fig. 21b) in form (the former form predominates). The average number of pigment granules in macrogametocytes is greater than 25, and it is approximately half as many in microgametocytes (compare Fig. 21a, b with Fig. 21c)
	………………………………………………………………………………………………………………….	*H. plataleae* (Fig. 21a–c) [3, 116]
2 (1)	Fully grown gametocytes usually do not occupy all or nearly all cytoplasmic space in infected erythrocytes, including poles of erythrocytes (Fig. 21d, f, i); they are thin halteridial in form (Fig. 21d–f, i). The average number of pigment granules in macrogametocytes is less than 25, and it is approximately the same in macrogametocytes and microgametocytes
3 (4)	Markedly ameboid forms predominate among young gametocytes (Fig. 21e). Advanced growing and fully grown gametocytes are predominantly markedly irregular in outline (Fig. 21d, f)
	………………………………………………………………………………………………………………….	*H. pelouroi* (Fig. 21d–f) [3, 116, 117]
4 (3)	Ameboid forms do not predominate among young gametocytes; outline of growing gametocytes is predominantly even (Fig. 21g–i). Advanced growing (Fig. 21g, h) and fully grown gametocytes (Fig. 21i) are predominantly even in outline, but gametocytes with slightly wavy margins might occur occasionally
	………………………………………………………………………………………………………………….	*H. herodiadis* (Fig. 21g–i) [3, 118]

Step	Features and species
1 (8)	Fully grown gametocytes do not encircle completely the nuclei of infected erythrocytes. Circumnuclear gametocytes (Fig. 22c, d) are absent
2 (11)	Fully grown gametocytes do not enucleate the infected erythrocytes
3 (16)	Clearly dumbbell-like or bilobed in shape (Fig. 22l, m) fully grown gametocytes are absent
4 (17)	The average number of pigment granules in fully grown macrogametocytes and microgametocytes is similar
5 (18)	Macrogametocyte nuclei are predominantly in median position (Fig. 22a, b). Fully grown gametocytes do not contain large compact volutin granules (Fig. 22p)
6 (7)	Advanced growing gametocytes and fully grown gametocytes are closely appressed to nuclei of infected erythrocytes (Fig. 22a)
	………………………………………………………………………………………………………………….	*H. xantholaemae* (Fig. 22a) [3, 119]
7 (6)	Advanced growing gametocytes and fully grown gametocytes, which do not touch nuclei of infected erythrocytes (Fig. 22b) are common
	………………………………………………………………………………………………………………….	*H. cornuata* (Fig. 22b) [3, 119]
8 (1)	Fully grown gametocytes encircle completely the nuclei of infected erythrocytes (Fig. 22c, d). Growing gametocytes readily tend to grow around nuclei of the erythrocytes
9 (10)	Advanced growing and fully grown gametocytes contain numerous prominent volutin granules and clearly distinct volutin clumps (Fig. 22c), which mask pigment granules^a
	………………………………………………………………………………………………………………….	*H. velans* (Fig. 22c) [3, 87, 120, 121]
10 (9)	Advanced growing and fully grown gametocytes do not contain numerous prominent volutin granules and their clearly distinct clumps (Fig. 22c), which mask pigment granules^a
	……………………………………………….	*H. homovelans* (Fig. 22d) [121]
11 (2)	Fully grown gametocytes enucleate the infected erythrocytes at final stage of development (Fig. 22e, g, h). Growing advanced gametocytes markedly displace nuclei of infected erythrocytes, often to erythrocyte poles (Fig. 22f, i, k)^b
12 (15)	The average number of pigment granules in fully grown gametocytes is less than 25
13 (14)	Infected erythrocytes are hypertrophied (on average approximately 10%) in length in comparison to uninfected ones
	………………………………………………………………………………………………………………….	*H. bennetti* (Fig. 22e–g) [3, 122]
14 (13)	Infected erythrocytes do not change significantly on average in length in comparison to uninfected ones
	………………………………………………………………………………………………………………….	*H. bucconis* (Fig. 22h) [3, 64]^c
15 (12)	The average number of pigment granules in fully grown gametocytes is greater than 25
	………………………………………………………………………………………………………………….	*H. thereicerycis* (Fig. 22i–k) [3, 119]
16 (3)	Clearly dumbbell-like or bilobed in shape (Fig. 22l, m) fully grown gametocytes are common
	………………………………………………………………………………………………………………….	*H. bilobata* (Fig. 22l, m) [3, 119]
17 (4)	The average number of pigment granules in fully grown macrogametocytes is approximately half as many as in fully grown microgametocytes (compare Fig. 22n and o)
	………………………………………………………………………………………………………………….	*H. indicator* (Fig. 22n, o) [3, 64]
18 (5)	Macrogametocyte nuclei are predominantly of sub-terminal position (Fig. 22p). Fully grown gametocytes contain large compact volutin granules
	………………………………………………………………………………………………………………….	*H. borgesi* (Fig. 22p) [3, 122]

PERMALINK

Keys to the avian Haemoproteus parasites (Haemosporida, Haemoproteidae)

Gediminas Valkiūnas

Tatjana A Iezhova

Abstract

Background

Methods

Results

Conclusion

Background

Methods

Results

Fig. 1.

Fig. 2.

Table 1.

Discussion

Fig. 3.

Fig. 4.

Fig. 5.

Fig. 6.

Fig. 7.

Fig. 8.

Fig. 9.

Fig. 10.

Fig. 11.

Fig. 12.

Fig. 13.

Fig. 14.

Fig. 15.

Fig. 16.

Fig. 17.

Fig. 18.

Fig. 19.

Fig. 20.

Fig. 21.

Fig. 22.

Fig. 23.

Fig. 24.

Fig. 25.

Fig. 26.

Fig. 27.

Fig. 28.

Fig. 29.

Fig. 30.

Fig. 31.

Fig. 32.

Fig. 33.

Fig. 34.

Fig. 35.

Fig. 36.

Fig. 37.

Fig. 38.

Table 36.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

Table 18.

Table 19.

Table 20.

Table 21.

Table 22.

Table 23.

Table 24.

Table 25.

Table 26.

Table 27.

Table 28.

Step	Features and species
1 (4)	Fully grown gametocytes are circumnuclear; they markedly enclose the nuclei of infected erythrocytes with their ends and finally can completely encircle the nuclei (Fig. 23a–c). The average number of pigment granules in gametocytes is greater than 15
2 (3)	Elongate (rod-like) pigment granules present and often predominate in advanced and fully grown gametocytes (Fig. 23a)
	…………………………………………………………………………………………………….	*H. handai* (Fig. 23a, b) [3, 123, 124]
3 (2)	Elongate (rod-like) pigment granules are rare, they do not predominate in advanced and fully grown gametocytes. Roundish or slightly oval pigment granules predominate (Fig. 23c)
	………………………………………………………………………….	*H. homohandai* (Fig. 23c) [125]
4 (1)	Fully grown gametocytes are halteridial, they do not encircle the nucleus of infected erythrocytes completely (Fig. 23d, e). The average number of pigment granules in gametocytes is less than 15
	………………………………………………………………………….	*H. psittaci* (Fig. 23d, e) [3, 126]

Step	Features and species^a
1 (2)	Nuclei of macrogametocytes are median in position (Fig. 24a)
	………………………………………………………………………………………………………………….	*H. krylovi* (Fig. 24a) [3]
2 (1)	Nuclei of macrogametocytes are subterminal in position^b
	………………………………………………………………………………………………………………….	*H. pteroclis* (Fig. 24b) [3, 127]

Step	Features and species
1 (2)	Volutin is conspicuous in fully grown gametocytes; it is arranged as compact roundish or circular granules, which tend to gather close to gametocyte ends (Fig. 25a-c). Fully grown gametocytes are halteridial (Fig. 25b) or close to circumnuclear, occasionally circumnuclear (Fig. 25c); halteridial forms (Fig. 25b) usually common
	………………………………………………………………………………………………………………….	*H. syrnii* (Fig. 25a–c) [3, 128–130]^b
2 (1)	Volutin is present, but is not conspicuous in fully grown gametocytes (Fig. 25d, e); it is dispersed more or less evenly (Fig. 25d, e), but not arranged as compact roundish or circular granules, which tend to gather close to gametocytes ends (Fig. 25a–c). Fully grown gametocytes are predominantly close to circumnuclear (Fig. 25d), often circumnuclear (Fig. 25e); halteridial forms (Fig. 25b) are rare
	………………………………………………………………………………………………………………….	*H. noctuae* (Fig. 25d, e) [3, 130]

Step	Features and species
1 (4)	Gametocytes tend to grow around nuclei of infected erythrocytes and advanced forms markedly enclose the nuclei with their ends. Fully grown gametocytes finally completely encircle the nuclei of infected erythrocytes (Fig. 27a–c) and can occupy all available cytoplasmic space in the erythrocytes (Fig. 27b)
2 (3)	Close to circumnuclear and circumnuclear gametocytes readily predominate among advanced and fully grown both macrogametocytes and microgametocytes (Fig. 27a, b). Halteridial gametocytes, which markedly displace nuclei of infected erythrocytes are absent or appear only occasionally
	………………………………………………………………………………………………………………….	*H. circumnuclearis* (Fig. 27a, b) [3, 133]
3 (2)	Circumnuclear gametocytes readily predominate among advanced and fully frown macrogametocytes (Fig. 27c), but do not predominate among advanced and fully frown microgametocytes, which often assume broadly halteridial forms and markedly displace host cell nuclei (Fig. 27d)
	………………………………………………………………………………………………………………….	*H. pittae* (Fig. 27c, d) [3, 134]
4 (1)	Gametocytes do not tend to grow around nuclei of infected erythrocytes, they do not markedly enclose the nuclei with their ends (Fig. 27e–i). Fully grown gametocytes are halteridial; they do not completely encircle nuclei of infected erythrocytes and never occupy all available cytoplasmic space in the erythrocytes (Fig. 27e, g–i)
5 (6)	Advanced growing gametocytes, which length is bigger than the length of erythrocyte nuclei, are closely appressed to the nuclei and envelope of infected erythrocytes (Fig. 27e)
	………………………………………………………………………………………………………………….	*H. tyranni* (Fig. 27e) [3, 133]
6 (5)	Advanced growing gametocytes, which length is bigger than the length of erythrocyte nuclei, predominantly do not touch the envelope of infected erythrocytes along their entire margins (Fig. 27f)
7 (8)	The average number of pigment granules in gametocytes is close to 10 (Fig. 27g). Nuclei assume predominantly subterminal position in macrogametocytes (Fig. 27f, g). Attenuated (snake-like) advanced gametocytes develop (Fig. 27f)
	………………………………………………………………………………………………………………….	*H. furnarius* (Fig. 27f, g) [3, 135]
8 (7)	The average number of pigment granules in gametocytes is close to 15 or greater (Fig. 27h, i). Macrogametocyte nuclei assume predominantly median position (Fig. 27i) or their position is variable (median, submedian and subterminal) in different cells (Fig. 27h, i). Attenuated (snake-like) advanced gametocytes (Fig. 27f) do not develop
9 (10)	The average number of pigment granules in fully grown gametocytes is close to 15; medium size pigment granules (0.5–1 µm) predominate, but large (greater than 1 µm) pigment granules also might occur (Fig. 27h)
	………………………………………………………………………………………………………………….	*H. formicarius* (Fig. 27h) [3, 135]
10 (9)	The average number of pigment granules in fully grown gametocytes is close to 20; small (usually less than 0.5 µm) pigment granules predominate, and large (greater than 1 µm) pigment granules do not occur (Fig. 27i)
	………………………………………………………………………………………………………………….	*H. eurylaimus* (Fig. 27i) [3, 134]

Step	Features and species
1 (5)	Growing advanced gametocytes (size greater than length of erythrocyte nuclei), which do not touch the envelope of infected erythrocytes along their entire margin (Fig. 28e, f), are absent
2 (6)	Nuclei predominantly assume subterminal or submedian position in fully grown macrogametocytes (Fig. 28a, b)
3 (4)	Fully grown macrogametocytes and microgametocytes are halteridial (Fig. 28b). Pigment granules are markedly variable in size and form in advanced gametocytes (Fig. 28a). Small (less than 0.5 μm) and medium (0.5 to 1.0 μm) size pigment granules occur approximately equally in fully grown gametocytes (Fig. 28a, b)
	………………………………………………………………………………………………………………….	*H. vireonis* (Fig. 28a, b) [3, 62, 135, 136]
4 (3)	Fully grown macrogametocytes are halteridial (Fig. 28d), but microgametocytes are more pleomorphic (halteridial, close to circumnuclear and occasionally even circumnuclear forms occur). Pigment granules are usually more or less uniform in size and form in advanced gametocytes (Fig. 28c, d). Pigment granules in fully grown gametocytes are usually of medium size (0.5 to 1.0 μm). Small (less than 0.5 μm) pigment granules are not present or are uncommon in fully grown gametocytes (Fig. 28d)
	………………………………………………………………………………………………………………….	*H. ptilotis* (Fig. 28c, d) [3, 137, 138]
5 (1)	Growing advanced gametocytes (size greater than length of erythrocyte nuclei), which do not touch the envelope of infected erythrocytes along their entire margin (Fig. 28e, f), are common
	………………………………………………………………………………………………………………….	*H. pachycephalus* (Fig. 28e, f) [3, 139]
6 (2)	Nuclei predominantly assume terminal or close to the terminal position in fully grown macrogametocytes (Fig. 28g, h)
	………………………………………………………………………………………………………………….	*H. orioli* (Fig. 28g, h) [3, 140]

Step	Features and species
1 (6)	Growing and fully grown macrogametocytes do not contain large (greater that 1.5 µm in diameter) vacuoles (Fig. 29d, e)
2 (7)	Fully grown gametocytes are halteridial; they reach poles of infected erythrocytes and occupy the poles completely (Fig. 29a)
3 (8)	Advanced growing gametocytes are closely appressed to nuclei of infected erythrocytes and often do not touch envelope of the erythrocytes along their entire margin (Fig. 29b, c)
4 (5)	The average number of pigment granules in fully grown gametocytes is greater than 12
	………………………………………………………………………………………………………………….	*H. aegithinae* (Fig. 29a–c) [3, 141]
5 (4)	The average number of pigment granules in fully grown gametocytes is less than 12
	………………………………………………………………………………………………………………….	*H. vangii*^a [142]
6 (1)	Advanced growing and fully grown macrogametocytes often contain one large (greater that 1.5 µm in diameter) vacuole, which is of sub-terminal position in gametocytes and adheres to the erythrocyte envelope; the vacuole usually locates at gametocyte end, which is opposite in location to the gametocyte nucleus (Fig. 29d, e). Occasionally, up to three or even more small vacuoles can be present
	………………………………………………………………………………………………………………….	*H. bukaka* (Fig. 29d, e) [143]
7 (2)	Fully grown gametocytes are microhalteridial; they usually do not reach poles of infected erythrocytes
	………………………………………………………………………………………………………………….	*H. madagascariensis*^a [142]
8 (3)	Advanced growing gametocytes often do not touch nuclei of infected erythrocytes along their entire margin, but are closely appressed to erythrocyte envelope (Fig. 29f)
	………………………………………………………………………………………………………………….	*H. cublae* (Fig. 29f) [3, 144]

Step	Features and species
1 (7)	The average number of pigment granules in fully grown gametocytes is less than 20. Circumnuclear gametocytes (Fig. 30e) are absent or occur only occasionally among fully grown macrogametocytes
2 (8)	The nuclei of macrogametocytes are in subterminal (Fig. 30a, b) or close to terminal (Fig. 30j) positions and, as a rule, they are not closely appressed to the nuclei of infected erythrocytes (Fig. 30g)
3 (9)	Dumbbell-shaped growing gametocytes (Fig. 30b, i) are common
4 (10)	Dumbbell-shaped growing gametocytes, which do not touch the envelope of erythrocytes along their entire margin (Fig. 30i), are absent. Fully grown gametocytes fill the infected erythrocytes up to their poles (Fig. 30a, d)
5 (6)	The cytoplasm is markedly vacuolated in growing and fully grown microgametocytes, with several distinct large (greater than 1 µm in diameter) vacuoles often present (Fig. 30c)
	………………………………………………………………………………………………………………….	*H. homopicae* (Fig. 30a–c) [100]
6 (5)	The cytoplasm is not vacuolated or only slightly vacuolated in growing and fully grown microgametocytes (Fig. 30d); it does not contain distinct large (greater than 1 µm in diameter) vacuoles (Fig. 30c)
	………………………………………………………………………………………………………………….	*H. picae* (Fig. 30d) [3, 87, 145]
7 (1)	The average number of pigment granules in fully grown gametocytes is 20 or greater. Circumnuclear gametocytes (Fig. 30e) are common among fully grown macrogametocytes
	………………………………………………………………………………………………………………….	*H. danilewskii* (Fig. 30e, f) [3, 118, 146]
8 (2)	The nuclei of macrogametocytes usually are in median (Fig. 30g) position and, as a rule, they are closely appressed to the nuclei of infected erythrocytes (Fig. 30g)
	………………………………………………………………………………………………………………….	*H. lanii* (Fig. 30g) [3, 23, 85, 118, 147–150]
9 (3)	Dumbbell-shaped growing gametocytes (Fig. 30b, i) are absent or appear only occasionally
	………………………………………………………………………………………………………………….	*H. dicruri* (Fig. 30h) [3, 151]
10 (4)	Dumbbell-shaped growing gametocytes, which do not touch the envelope of erythrocytes along their entire margin (Fig. 30i), are common and predominate. Fully grown gametocytes usually do not fill the infected erythrocytes up to their poles (Fig. 30j)
	………………………………………………………………………………………………………………….	*H. monarchus* (Fig. 30i, j) [3, 139]

Step	Features and species
1 (8)	Roundish fully grown gametocytes (Fig. 31e, f) are absent
2 (9)	The nuclei of macrogametocytes usually are in subterminal position and do not locate close to the erythrocyte nuclei (Fig. 31a, b, d, i)
3 (10)	Fully grown gametocytes, which are closely appressed to the nuclei of erythrocytes but do not touch the envelope of the erythrocytes along their entire margin (Fig. 31h, i), are absent
4 (5)	The average number of pigment granules in fully grown gametocytes is greater than 15 (Fig. 31a)
	………………………………………………………………………………………………………………….	*H. wenyoni* (Fig. 31a) [3, 139]
5 (4)	The average number of pigment granules in fully grown gametocytes is less than 15 (Fig. 31b)
6 (7)	Dumbbell-shaped gametocytes are present among growing advanced macrogametocytes (Fig. 31c), and such forms are common
	………………………………………………………………………………………………………………….	*H. majoris*^a (Fig. 31b, c) [3, 28, 43, 91, 152, 153]
7 (6)	Dumbbell-shape gametocytes (Fig. 31c) are absent
	………………………………………………………………………………………………………………….	*H. alaudae* (Fig. 31d) [3, 141]
8 (1)	Fully grown gametocytes are roundish; they markedly deform infected erythrocytes, markedly displace their nuclei, and can even enucleate the host cells (Fig. 31e, f)
	………………………………………………………………………………………………………………….	*H. parus*^b (Fig. 31e, f) [3, 154]
9 (2)	The nuclei of macrogametocytes are median or submedian in positions; they usually adhere to the erythrocyte nuclei (Fig. 31g)
	………………………………………………………………………………………………………………….	*H. calandrellae* (Fig. 31g) [3, 155]
10 (3)	Fully grown gametocytes are closely appressed to the nuclei of erythrocytes but do not touch the envelope of the erythrocytes along their entire margin (Fig. 31h, i)
	………………………………………………………………………………………………………………….	*H. nucleophilus* (Fig. 31h, i) [3, 156]

Step	Features and species
1 (10)	Gametocytes do not contain gigantic pigment granules (greater than 1.5 μm) (Fig. 32k, l). The average number of pigment granules in fully grown gametocytes is greater than five
2 (5)	The nuclear material is condensed in fully grown microgametocytes (Fig. 32b, c); the size of the nuclei in microgametocytes does not exceed that of the nuclei in macrogametocytes (compare Fig. 32a and Fig. 32b)
3 (4)	Fully grown gametocytes are microhalteridial; they usually do not reach poles of infected erythrocytes (Fig. 32a, b). Growing advanced gametocytes, which do not touch envelope of infected erythrocytes along their entire margins (Fig. 32d), are absent. Pigment granules in fully grown gametocytes tend to clamp in one or two louse groups, which often locate close to the gametocyte ends (Fig. 32b), however, the gametocytes with scattered pigment granules are also present (Fig. 32a)
	………………………………………………………………………………………………………………….	*H. payevskyi* (Fig. 32a, b) [3, 147, 148, 157]
4 (3)	Fully grown gametocytes are halteridial (Fig. 32c); they usually reach poles of infected erythrocytes (Fig. 32c). Advanced growing gametocytes, which do not touch envelope of infected erythrocytes along their entire margins, are present (Fig. 32d). Pigment granules in fully grown gametocytes are usually scattered (Fig. 32c, d), they do not tend to clamp in one or two louse groups, which locate close to the gametocyte ends (Fig. 32b)
	………………………………………………………………………………………………………………….	*H. nucleocondensus* (Fig. 32c, d) [148, 149, 158]
5 (2)	The nuclear material is diffuse (not condensed) in fully grown microgametocytes (Fig. 32e, j); the size of the nuclei in microgametocytes markedly exceeds that of the nucleus in macrogametocytes (compare Fig. 32e and Fig. 32f)
6 (7)	Gametocytes grow around nuclei of infected erythrocytes and only slightly (if at all) displace the nuclei laterally (Fig. 32g). Fully grown gametocytes markedly enclose the erythrocyte nuclei with their ends and are predominantly close to circumnuclear in form (Fig. 32f); completely circumnuclear forms might occur (Fig. 32e), but do not predominate. Ends of advanced growing gametocytes predominantly are more or less irregular or ameboid in outline (Fig. 32g)
	………………………………………………………………………………………………………………….	*H. belopolskyi* (Fig. 32e–g) [3, 147–149, 159, 160]
7 (6)	Gametocytes grow along nuclei of infected erythrocytes and markedly displace the nuclei laterally (Fig. 32h–j). Fully grown gametocytes are broadly halteridial (Fig. 32h, j); they only slightly enclose the erythrocyte nuclei with their ends (Fig. 32h, j); circumnuclear or close to circumnuclear gametocytes (Fig. 32e, f) are absent. Ends of advanced growing gametocytes predominantly are more or less even in outline (Fig. 32i)
8 (9)	The nuclei of fully grown macrogametocytes are predominantly median or submedian in position (Fig. 32h). Fully grown gametocytes contain the pigment granules of more or less uniform size, and oval to slightly elongate granules of medium size (0.5–1 μm) predominate (Fig. 32h). Large (1–1.5 μm) pigment granules are absent
	………………………………………………………………………………………………………………….	*H. parahirundinis* (Fig. 32h) [161]
9 (8)	The nuclei of fully grown macrogametocytes are subterminal in position (Fig. 32i). Fully grown gametocytes contain pigment granules of markedly variable size, and small (< 0.5 μm), medium (0.5–1 μm) and sometimes even large (1–1.5 μm) pigment granules might occur simultaneously in same gametocytes (Fig. 32i, j)
	………………………………………………………………………………………………………………….	*H. hirundinis* (Fig. 32i, j) [3, 149, 161]
10 (1)	Gigantic pigment granules are present in fully grown gametocytes (Fig. 32k, l). The average number of pigment granules in fully grown gametocytes is less than five
	………………………………………………………………………………………………………………….	*H. stellaris* (Fig. 32k, l) [3, 161, 162]

Step	Features and species
1 (12)	Advanced dumbbell-shaped gametocytes, which do not touch envelope of infected erythrocytes along their entire margin and have highly constricted (attenuated) central portion and readily thickened roundish ends (Fig. 33j), are absent
2 (13)	The majority of advanced macrogametocytes do not contain one clear roundish discrete vacuole, which is close to 1 µm in diameter (Fig. 33k, l)
3 (16)	Fully grown gametocytes are halteridial (Fig. 33a, f, h, g) or close to circumnuclear (Fig. 33c–e); they reach poles of infected erythrocytes and usually occupy poles of the erythrocytes completely (Fig. 33a, c–h)
4 (17)	Advanced growing macrogametocytes, which are closely appressed to the nuclei of infected erythrocytes but do not touch the envelope of the erythrocytes along their entire margin (Fig. 33n), are absent. Advanced growing gametocytes touch envelope of infected erythrocytes entirely (Fig. 33p) or in one/several points (Fig. 33b)
5 (20)	Advanced growing gametocytes, whose pellicle in the centre does not extend to the envelope of infected erythrocytes causing a ‘dip’ and giving a dumbbell-like appearance (Fig. 33b), are present and usually common
6 (7)	Fully grown gametocytes contain large (1–1.5 µm) elongate pigment granules (Fig. 33a), which are common and might predominate in some cells
	………………………………………………………………………………………………………………….	*H. killangoi* (Fig. 33a, b) [3, 93, 163, 164]
7 (6)	Fully grown gametocytes usually do not contain large (1–1.5 µm) elongate pigment granules (Fig. 33a); such pigment granules can occur only occasionally in some cells
8 (9)	Fully grown gametocytes markedly enclose nuclei of infected erythrocytes with their end and tend to assume circumnuclear form (Fig. 33c–e); circumnuclear gametocytes sometimes occur (Fig. 33e), more often among microgametocytes. Fully grown gametocytes are predominantly close to circumnuclear in form (Fig. 33d)
	………………………………………………………………………………………………………………….	*H. parabelopolskyi*^b (Fig. 33c–e) [100, 128, 148, 159]
9 (8)	Fully grown gametocytes slightly enclose nuclei of infected erythrocytes with their ends (Fig. 33f–h), but do not tend to assume circumnuclear form. Fully grown gametocytes are predominantly halteridial (Fig. 33f, g, h)
10 (11)	Growing and fully grown gametocytes markedly displace nuclei of infected erythrocytes laterally (Fig. 33f, g). Fully grown gametocytes often displace erythrocyte nuclei close to the erythrocyte envelope (Fig. 33g). The average NDR is less than 0.7
	………………………………………………………………………………………………………………….	*H. leiothrichus, H. homoleiothrichus*^c,d (Fig. 33f, g) [7, 165]
11 (10)	Growing and fully grown gametocytes do not markedly displace nuclei of infected erythrocytes laterally (Fig. 33h). Fully grown gametocytes usually do not displace erythrocyte nuclei close to the erythrocyte envelope (Fig. 33g). The average NDR is 0.7 or greater. Early gametocytes often are markedly ameboid in outline (Fig. 33i)
	………………………………………………………………………………………………………………….	*H. timalus*^d (Fig. 33h, i) [3, 139]
12 (1)	Advanced dumbbell-shaped gametocytes, which do not touch envelope of infected erythrocytes along their entire margin and have highly constricted (attenuated) central portion and readily thickened roundish ends (Fig. 33j), are common
	………………………………………………………………………………………………………………….	*H. philippinensis* (Fig. 33j) [3, 166]
13 (2)	The majority of advanced macrogametocytes contain a clear roundish discrete vacuole, which is close to 1 µm in diameter (Fig. 33k, l)
14 (15)	The average number of pigment granules in fully grown gametocytes is greater than 15
	………………………………………………………………………………………………………………….	*H. vacuolatus* (Fig. 33k) [63, 100, 167, 168]
15 (14)	The average number of pigment granules in fully grown gametocytes is less than 15
	………………………………………………………………………………………………………………….	*H. palloris* (Fig. 33l) [63, 91, 100, 168]
16 (3)	Fully grown gametocytes are microhalteridial, they do not reach poles of infected erythrocytes and do not occupy poles of the erythrocytes (Fig. 33m)
	………………………………………………………………………………………………………………….	*H. homogeneae* (Fig. 33m) [100]
17 (4)	Advanced growing macrogametocytes, which are closely appressed to the nuclei of infected erythrocytes but do not touch the envelope of the erythrocytes along their entire margin (Fig. 33n), are common
18 (19)	Nuclei of fully grown macrogametocytes are subterminal in position (Fig. 33)
	………………………………………………………………………………………………………………….	*H. otocompsae* (Fig. 33n, o) [3, 166]
19 (18)	Nuclei of fully grown macrogametocytes are median in position (Fig. 33p)
	………………………………………………………………………………………………………………….	*H. sanguinis* (Fig. 33p) [3, 166]
20 (5)	Advanced growing gametocytes, whose pellicle in the centre does not extend to the envelope of infected erythrocytes causing a ‘dip’ and giving a dumbbell-like appearance (Fig. 33b), are absent or appear only occasionally. Advanced growing gametocytes are closely appressed both to the nuclei and envelop of infected erythrocytes (Fig. 33v)
21 (24)	Fully grown gametocytes do not contain large (greater than 1 µm) pigment granules (Fig. 33t–v). The macrogametocyte cytoplasm is relatively pale-stained and is similar to microgametocytes based this character (compare Fig. 33q and r)
22 (23)	Fully grown gametocytes contain roundish, small (less than 0.5 µm) pigment granules (Fig. 33q, r). Pigment granules do not change size and shape significantly as parasite matures
	………………………………………………………………………………………………………………….	*H. pallidulus* (Fig. 33q, r) [63, 100, 168, 169]
23 (22)	Fully grown gametocytes contain roundish and oval pigment granules (Fig. 33s), which are of medium size (0.5–1.0 μm). Pigment granules readily increase in size as parasite matures
	………………………………………………………………………………………………………………….	*H. homopalloris* (Fig. 33s) [63, 168]
24 (21)	Fully grown gametocytes contain large (greater than 1 µm) pigment granules (Fig. 33t–v). Macrogametocytes are readily distinguishable from microgametocytes based on intensity of staining of their cytoplasm (Fig. 33t, u)
	………………………………………………………………………………………………………………….	*H. zosteropis* (Fig. 33t–v) [3, 163, 164]