Measurements of blue label uptake after 30 minutes of feeding and the proportion of feeding events observed during this period, where one circle represents one vial containing 5 flies.

<p>Trend lines represent the relationship between the volume of food ingested and the observed proportion of flies feeding (V/O) described in <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0006063#pone-0006063-t001" target="_blank">Table 1</a>. Dashed lines represent open circles. All flies were female unless stated, were 7 days old and were allowed to mate for 48 hours after eclosion (NF = the number of flies per condition, NV = the number of vials per condition). (a) A linear (V/O) relationship existed in mated Dahomey females (NF = 210, NV = 42). (b) The V/O relationships of mated Dahomey males and females did not differ significantly, although females were found to have fed at a greater frequency than males during the 30 minutes (NF = 200, NV = 40). The gradient for males did not differ significantly from that for females but had a lower intercept. (c) DR fed and full fed Dahomey females shared the same V/O relationship and no difference in feeding between dietary conditions was found with the combined assay (NF = 75, NV = 15). (d) The V/O relationship was the same in <i>chico</i>¹ heterozygotes and in the Dahomey control. No difference in feeding between genotypes was found with the combined assay (NF = 90, NV = 18). (e) The V/O relationship was the same in <i>takeout </i>¹ and in Canton-S females, even though <i>takeout </i>¹ flies were found to feed at a higher frequency than Canton-S controls (NF = 60, NV = 12). (f) Both <i>ovo</i>^D1 and <i>white</i>^Dahomey females had a positive V/O relationship, but <i>ovo</i>^D1 flies had a significantly greater gradient and intercept, and therefore increased the volume of food ingested per proboscis-extension more quickly than <i>white</i>^Dahomey females (NF = 200, NV = 40).