226 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 



Third staf/c. — Length, V2-lo mm. The charactei's of the liinil stage now appear. The head 

 has changed its shape and style of inarkiugs to that of the last stage; it is flatter in front, with 

 a lateral brown line edged with white, wliile the large, conspicnoas, dark spots have disappeared,, 

 and the color of the head is dull opaque-amber. The four red, parallel, dorsal lines on the second 

 and third thoracic and seventh and eighth abdominal segments are now distinct. All the dorsal 

 tubercles except those on tlic prothoracic and first and eighth abdominal segments have muck 

 diminished in size, while the others, have remained stationary. 



Fourth staije. — Length, 18 mm. The piliferons warts in general are smaller than in Stage III, 

 and those on the prothoracic and lirst and eighth abdominal segments are snniller tlian before. 

 The eighth abdominal segment is slightly humped, and the anal legs are normal, though about 

 one-half as thick as those in front. The body is green, with a broad subdorsal and two nai-row 

 lateral yellow lines, as in the last stage, the lower being the infra-si)iracalar line. The sides of the 

 three thoracic segments are dotted with reddish i)ink, and there is a reddish stieak on the outside 

 of the anal legs. The subdorsal yellow lines diverge on the prothoracic segment, and along the 

 next two segments succeeding are edged within with pink red lines. Behind the two dorsal' 

 tubercles on the first abdominal segment they are much fartlier a])art, extending to the supra-anal 

 plate, and are whitish yellow, narrowly bordered with deep, straw-yellow, and inclose a narrow, 

 yellow dorsal line. (This line in the next stage extends to the prothoracic segment.) 



Fifth and last staijc. — Lengtli, 30-32 mm. It differs in the dorsal piliferons warts on the first 

 thoracic and first and eighth abdominal segments being smaller than in the fourth stage, being 

 now no larger than those on the other segments, and the humj) on the eightli segment has almost 

 disappeared. There is, as in the fourth stage, a conspicuous red dash on each side of the third 

 abdominal segment, and the other lines are as described in the fourth stage. 



RECAPITULATION. 



1. Head large, with dark spots and connected lines in Stages I and IT. 



-'. The spots disappear, and the peculiar lateral dark line edged with white characteristic of 

 the final stage appears in Stage III. 



3. The piliferons tubercles on first thoracic and first and eighth abdominal segments attain 

 their maximum in Stage II; the tendency after this stage is to return to a simple, smooth body, 

 without excessive ornamentation or any decided change in coloration. 



4. In Stage III all the other tubercles diminish in size. 



5. The style of coloration of Stage V is indicated in Stage II. 



G. In Stage IV the tubercles almost reach their minimum, becomin-g still smaller in the final 

 stage. 



7. The few tenant hairs present in the first stage are battledoor-shaped. 



It is interesting to notice, in reviewing the larval history of this species, the strong tendency 

 shown after the second stage to a diminution in size of the tubercles, so that by the fourth stage 

 the body becomes smooth and free from all projections, humps, and spines, and thus more 

 uoctuiform. At the same time the yellow and whitish stripes and pink blotches become indicated 

 at an earlier stage than usual, as if the aim were to adapt the caterpillar to the ribs and parallel 

 greenish and yellowish lines or shades of the leaf on which it feeds. 



This is perhaps, as regards the other species, the most generalized and simple form in its 

 early larval stages, there being no horns and the dorsal warts of nearly uniform size. 



lu the group comprising H. biundata, f/uttiritta, and obliqun there is a singular degree of liyper- 

 trophy and specialization of the dorsal tubercles, while in the group represented by i/. HH/oti/or 

 the process of hypertrophy and specialization takes another direction, i. e., the anal legs, the larvie- 

 becoming cerura-forra. 



Eemakks.— PI. XXIX, figs. 2, 2rt represent what may prove to be the young larva of this spe- 

 cies. It was found by Mr. Bridgham on the walnut at Providence, It. I., July 5. I have no notes, 

 on it. The following descriptions have been drawn up from Oomstock's specimens, the types of his. 

 description in his report as United States Entoaiologist for ISSd. His No. 21!», "Notodonta ou 

 oak, September 23, 1879," is equal to var. c. of his description. I am indebted to Professor liiley 



