TRANSFORMATIONS OF EUM^US ATALA. 419 



tion is derived from the structure of tlie Euma^ida?, which, while they must be classed with 

 the latter of the two groups, give strong evidence of their affinities to the former. The 

 onisciform or sub-onisciform character of the larvae in these two large groups, their retrac- 

 tile or semi-retractile heads, the compact form of the chrysalids and their closely girt attach- 

 ment to their support, together with the opposite development of the forelegs in the 

 two sexes of the imago and the close similarity of all other points in the structure of the 

 perfect form show that no family distinction can be drawn between the two groups; 

 while the characters they have in common are very many of them in striking contrast to 

 those found in any other family of butterflies. It is perhaps in the larvae that the two 

 groups show the greatest distinction from each other, and this distinction is bridged over by 

 the aberrant tribe of Eumseidoe, leading us to suppose that were the transformations of 

 other allied forms known, we should find many more points of similarity than at present. 



Eumceus Atala is an exceedingly common insect in Cuba, and is found in great abundance 

 on the Florida Keys and the southern portion of the peninsula itself. Mr. C. J. Maynard, 

 who has travelled extensively over that portion of Florida, tells me that it hardly occurs 

 north of Miami on the east coast, although the food-plant of the larva extends north as far 

 as Smyrna. 



In Florida the butterfly emerges from the chrysalis early in January and is abundant 

 throughout this month. I found it also abundant in Cuba late in February or early as 

 March and took specimens at that time emerging from the chrysalis. Dr. Palmer saw the 

 butterfly laying her eggs at Biscayne Key in July, and during this month caterpillars of 

 every stage were found. The caterpillar feeds upon both surfaces of the leaves of Zamia 

 mterjr [folia, but seems to prefer the upper surface. The chrysalis is usually attached to the 

 under surface of the leaves or to the stem of the same plant, but sometimes to other plants 

 and even to grass, the head uppermost. 



Explanation of Plate xiv. 



(Figs. 1 and 2 bi-'long to the preceding memoir.) 



Fig. 3. Penultimate stage of caterpillar ; side view. 



Fig. 4. Tlie same ; side view of one of the abdominal segments, enlarged. 



Fig. 5. The same ; dorsal view. 



Fig. 6. Final stage of caterpillar; front view of head and first two thoracic segments, enlarged, showing 

 the tumescence of the first thoracic segment and the ventral expansion of the caterpillar at this stage, in 

 contradistinction to its cylindrical form in the previous stages ; the latter feature is less conspicuous on the 

 plate than in nature. 



Fig. 7. The same ; side view of parts of three abdominal segments, enlarged. 



Fig. 8. The same ; side view ; the artist has given much too cylindrical an appearance to the catei-jiillar. 



Fig. 9. The same ; dorsal view, the head concealed by the overhanging first thoracic segment. 



Fig. 10. Chrysalis ; side view with attachment. 



Fig. 11. The same; veiitral aspect. 



Fig. 12. The same ; dorsal aspect. 



Fig. 13. Imago ; neuration of fore wing of female, enlarged. 



Fig. 14. Tlie same; male; the left side shows the upper, the right the under surface. 



Fig. 15. The same ; neuration of extremity of cell of fore wing in the female, enlarged. 



Fig.16. The same; fore tarsi and part of tibia of female, enlarged. 



Fig. 17. The same ; fore tarsi and part of tibia of male, enlarged. 



N. B. — The drawings of the earlier stages were all taken from specimens preserved in alcohol. Figs. 13, 

 15, 16, 17 were drawn by S. H. Scudder; the remainder by S. E. Cassino. 



