Table 1.

Effect of C. bombi infection (yes/no), flowering strip treatment (high-infection wildflower, low-infection wildflower, or canola only) and their interaction (all fixed effects) on multiple measures of B. impatiens microcolony performance using generalized linear mixed models

	Surviving workers		Number of larvae		Larval weight		Number of eggs		Egg weight
	χ2	P	χ2	P	χ2	P	χ2	P	χ2	P
Infection	0.04	0.845	0.97	0.326	0.01	0.930	3.01	0.083	1.12	0.289
Treatment	15.77	<0.001	13.70	0.001	0.84	0.657	2.79	0.248	10.72	0.005
Inf × treat	3.32	0.190	2.22	0.33	5.03	0.081	0.37	0.833	0.65	0.722
With nectar as a covariate:
Nectar	15.39	<0.001	5.84	0.016	0.95	0.330	7.68	0.021	7.23	0.007
Infection	0.28	0.600	1.81	0.178	0.01	0.940	3.26	0.071	1.47	0.225
Treatment	3.34	0.188	5.92	0.052	4.97	0.083	0.06	0.97	3.49	0.175
Inf × treat	3.84	0.146	2.42	0.298	5.95	0.051	0.41	0.816	0.66	0.719

Surviving workers, number of larvae, and number of eggs were analyzed with Poisson distributions, and larval and egg weight with Gaussian distributions. The lower values show the same analysis with available nectar resources per tent included as a covariate. Bold indicates P < 0.05; italics indicate P < 0.055.