Extensive parental care experience of male seahorses increases their future mating attractiveness

Male-only parental care, where the male is the sole carer of his offspring, is a rare strategy in most animals. In mammals, the majority of families (about 90%) show female-only care and male-only care is absent. In birds, biparental care is the norm (about 90% of families) and male-only care is present in fewer than 2% of families. In reptiles, about 85% of families show no care, but of those that do, female-only care is dominant and there is also no male-only care (Reynolds et al. 2002). While in teleost fish, similar to the reptiles, most families (about 78%) do not provide care, but among those providing care, male acting as the sole carer accounts for more than 50% (Gross and Sargent 1985; Gross 2005). Why male-only care should be so prevalent in fish, including in species where sexual selection is stronger in males than females, has attracted much attention. The most common explanation seems to come from the application of Williams’ principle: the trade-off between current and residual reproductive values. The explanation holds that the prevalence of male-only care in fish is due to that male fish may pay a lower future cost than females for the same care benefits (Gross 2005). In addition, a number of studies have found that male fish with previous care experience are preferred by females (Goldberg et al. 2020). The finding of care-experienced males being favored by females suggests that the prevalence of male-only care in fish may also be explained by sexual selection through female preference for caring males (Lindström et al. 2006; Lindström and St. Mary 2008). The finding has been confirmed in a few fish, such as the 15-spined stickleback Spinachia spinachia (Östlund and Ahnesjö 1998) and the sand goby Pomatoschistus minutus (Lehtonen and Lindström 2007). However, all of these species provide a similar type of paternal care, which includes fanning, cleaning, and protecting the eggs in a purpose-build nest. It is still unclear whether the finding is universal across the other types of paternal care, such as those that provide nutrition to offspring. Seahorses (genus Hippocampus, family Syngnathidae) are also fish species showing male-only care. The care way of seahorses is mainly in incubating eggs. The female seahorses deposit eggs into the male seahorses’ brood pouch, where the eggs are fertilized. Then, male seahorses provide care for embryos, including protecting, aeration, nourishing, and osmoregulation, before releasing them as independent young (Stölting and Wilson 2007). Male care in seahorses, often called male pregnancy, is unique among the care-giving fish. In the present study, we used the lined seahorse Hippocampus erectus as a model of provision nutrition to offspring to investigate the relationship between male’s parental care experience (measured by the number of previous pregnancies) and his subsequent mating attractiveness (measured by the probability of being selected as the mate by females). The aim is to test whether the care type like seahorses also helps to increase the male’s future mating attractiveness, thereby to get a better understanding of male-only care in fish. Two experiments were performed in the present study. Experiment 1 tested female preference for 2 unfamiliar males, of which one was a virgin male with no pregnancy experience, and the other one was a mated male with pregnancy experience. Experiment 2 tested female preference for 2 familiar males. These 2 males differed in their initial preference and previous pregnancy experience. One was initially non-preferred by a focal female, but then he mated 3 times with the female (i.e., the initially non-preferred male had 3 pregnancy experiences), and the other one was initially preferred by the focal female, but then he mated only once with the female (i.e., the initially preferred male had only 1 pregnancy experience). These 2 experiments were independent and there was no reused individual between them. The seahorses were all virgins before experiments. Detailed methods are described in Supplementary Material 1.

A total of 58 replicates were tested in Experiment 1, of which 44 replicates where the female selected the mated male and 14 replicates where the female selected the virgin male (Table 1). Binomial test shows that females highly significantly preferred the mated males (P < 0.001). A total of 45 replicates were tested in Experiment 2, of which 13 replicates where the female selected the initially preferred male which had only 1 pregnancy experience and 32 replicates where the female selected the initially non-preferred male which had 3 pregnancy experiences (Table 2). Binomial test shows that females highly significantly preferred the males with more pregnancy experiences (P = 0.007), although they were initially non-preferred.

Table 1.

Results of female preference for the virgin male and the mated male in Experiment 1

	Total replicates (=P + Q+R)	P (=S + T)	Q	R	S	T	N (=Q + S)
Batch 1	25	17	5	3	13	4	18
Batch 2	25	18	4	3	15	3	19
Batch 3	25	19	5	1	16	3	21
Sum			14		44		58

Notes: P, the number of replicates where the female selected the mated male among the total of 25 replicates. These replicates continued the next step to verify their virgin males’ fertility. Detailed methods were described in Supplementary Material 1. Q, the number of replicates where the female selected the virgin male among the total of 25 replicates. R, the number of replicates with no mating among the total of 25 replicates. S, the number of replicates where virgin male succeeded in mating among the P replicates. T, the number of replicates where virgin male failed in mating among the P replicates. N, the number of valid replicates.

Table 2.

Results of female preference for the initially preferred male with 1 pregnancy and the initially non-preferred male with 3 pregnancies in Experiment 2

	W (=X + Y+Z)	X	Y	Z	N (=X + Y)
Batch 1	15	4	9	2	13
Batch 2	20	6	11	3	17
Batch 3	17	3	12	2	15
Sum		13	32		45

Notes: W, the total number of replicates. X, the number of replicates where the female selected the preferred male among the W replicates. Y, the number of replicates where the female selected the non-preferred male among the W replicates. Z, the number of replicates with no mating among the W replicates. N, the number of valid replicates.

According to the methods described in Supplementary Material 1, in the present study, the mated males and the initially non-preferred males being favored by the females results from several possible causes. For example, these 2 types of males both had just given birth and females might be attracted by the postpartum pheromones released by the males shortly after parturition. Another situation, the initially non-preferred males stayed together with females longer, and females might prefer the males who spent more time with them. However, based on our previous study, the possible causes mentioned above should not make sense, instead, that females might prefer the males with extensive experiences is the substantial cause because the 2 types of males both had more pregnancy experiences than their tank companions. In other words, the winning out of mated males and initially non-preferred males from their tank companions is mainly due to their advantages in pregnancy experiences. Detailed explanations for possible causes and refution are presented in Supplementary Material 2. Therefore, the main conclusion of the present study is that male seahorses with more pregnancy experiences are preferred by females. Similar to the care type like sand goby and 15-spined stickleback, the care type like seahorse can also increase a male’s future mating attractiveness. These findings not only help us better understand male-only care in fish but also help reveal that male-only care may have evolved as a result of sexual selection rather than natural selection alone (Kvarnemo 2006; Lindström and St. Mary 2008). In addition, the finding of the present study may also provide an insight into the mating system of seahorses. Seahorses show reproductive monogamy. The bonds of paired seahorses are stable and they can last through multiple breeding cycles, even years. The paired seahorses rarely change mate unless one member of the pair has a health problem or dies. The finding that males increase the mating attractiveness to their current mates through the accumulation of the number of pregnancy (mating) could probably explain why the pair bond is stable in seahorses. For male seahorses, they need to mate as many as possible in a limited breeding period to increase their mating attractiveness. Keeping with the same mate allows the breeding period to be used for mating as many as possible, because it not only saves time for finding a new mate but also saves time for getting familiar each other when mating with a new mate. As for why female seahorses prefer more pregnancy-experienced males, we believe that these males have a larger brood pouch opening may be one of the causes. Male seahorse’s pouch opening becomes larger with the increase of pregnancy number (personal observation), that is, the experienced males usually have a larger pouch opening which makes them easier to receive eggs during mating. A large opening could make mating success in a few seconds, while males with a small opening would take a few minutes to complete mating, even fail in mating (personal observation). That is why the males with a larger pouch opening are attractive to the females. In addition, other causes should not be excluded, such as the more mating experienced males might produce more or better offspring, or the more mating experienced males would have a shorter gestation period. All these will be investigated in the future.

Funding

Financial support was provided by the National Natural Sciences Foundation of China (Grant 32172950), the Natural Sciences Foundation of Shanghai (Grant 21ZR1479800), and the Central Public-interest Scientific Institution Basal Research Fund of CAFS (Grants 2017HYZD0401 and 2020TD53).

Conflict of Interest

All authors declare no conflict of interest.