THE CANADIAN ENTOMOLOGIST. 61 



second ; head, thorax and abdomen above strongly metalh'c, the head and 

 thorax bhiish, with tlie hair black, the abdomen yellowish-green, but bluish 

 apically, the tuft dark steel-blue ; patagia with a white spot ; neck with an 

 elongate crimson mark on each side ; front with a large transversely oval 

 white spot ; orbits margined with white above and in front, broadly at 

 sides of face ; the large anterior coxju broadly white in front ; a large white 

 spot on each side at base of abdomen ; anterior wings above a sub- 

 metallic blue-black, the basal and apical field not at all differently 

 coloured ; posterior wings with the upper half normally overlapped by the 

 anterior wings, shining whitish ; beneath, anterior wings are white where 

 they overlap the posterior, and are otherwise distinctly more metallic than 

 above. 



Among bananas at Boulder, Colorado, doubtless imported from 

 Central America. 



A similar, perhaps identical, insect, from Honduras, is in the National 

 Museum, as I learn from Dr. Dyar. The specific name is from Musa, 

 the banana. 



The type will be sent to U. S. National Museum. 



The C. Btitleri group may be tabulated thus : 



Neck with crimson spots i. 



Neck without crimson spots x. 



1. Basal half of anterior wings shining green viridis, Druce. 



Basal half of anterior wings coloured like the rest 2. 



2. Front with a large white spot musicohi, Ckll. 



Front without a white spot (S. America) Butieri, Moschl. 



3. Front with a white spot British Columbia specimen. 



Front without a white spot (Venezuela) caurensis, Klages, 



Dr. Dyar reports that Mr. Cockle's specimen lias the basal half of 

 fore wings shining green, as Druce describes for viridis, but it has no red 

 spots on thtr neck, while it has a white spot on the front. Thus it is near 

 to viridis, but not the same. 



The whole series affords a very good example of "Kaleidoscopic 

 variation," with different combinations of the same unit characters. 

 ^Vhether or not these forms are fixed in nature, no doubt they tould 

 easily be obtained pure and constant by a breeder, following Mendelian 

 methods. 



