27 



names of oss//'ra^i^^us, cain-i'tius, corax fruicifalis, or perhaps even 

 cryptoleuciis. The variation between early and late sets is very great in 

 some, and small in other sections of the country. However, some dif- 

 ference does exist, and always in one way. The eggs in the earlv sets av- 

 erage .06x03 larger than those of the late sets. The latter are not known 

 to be second sets, but are believed to be those of the more timid and 

 wary birds, building after the leaves have yielded a more effectual screen 

 for their nests. While I have no conclusive proof of it, I believe the 

 earlier sets are those of the older, or more hardy, or better developed 

 birds. It is remarkable that the latter not only produce the larger egg, 

 but also the greater number in a set. In five instances where it is known 

 that the birds laid second sets after the first had been taken, the second 

 sets averaged .02x01 less than the former, and usually cfintained one or 

 two eggs less in a clutch. 



The average measurements of 979 eggs is 1.65x1.15. This average is 

 considerably under that of our professional brethren : vet in all proba- 

 bility no average measurement of the eggs of this species is founded 

 on as numerous and accurate data as is that just given. 



It has often been a question to many whether the number of eggs in a 

 clutch has any influence whatever over the size of the egg. It is often 

 remarked that the larger the clutch the smaller the egg, and i<ice I'ei-sa. 

 This is true to a certain extent ; there are, of course, exceptions to all 

 rules ; for instance ; a set of seven eggs collected in Wisconsin gives an 

 average of 1.69x1. 19, which is slightly larger than the true average for that 

 part of the country. A series of sets of seven (if they could be had) 

 might show a very different average. 



The following tabulated form will render this more clear. 



Average size of 21 sets of 6 eggs each, 1.66 x 1.16 



<So " 5 " 1.65x1.15 



49 " 4 " 1 . 66 X 1 . 1 8 



16 "3 and 2 " 1.61 \ 1. 1 ^ 



It w'ill be at once seen that the sets of four average largest, and that 

 the sets of five average precisely the same as the total average of all the 

 eggs measured (this number is without doubt the most normal clutch of 

 americamis). The slight increase in the average of the sets of six can 

 readily be explained. In the first place it has been shown that the sets of 

 this number of eggs were all collected very early in the serison, and that 

 the supposition is that they were from the most hardy and vigorous birds ; 

 and again such clutches are more frequent in the North and West, where 

 all eggs average larger. The sets of two and three eggs each, which were 

 mostlv collected late in the season, are probably second or third chitches 



