MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 239' 



broad white, dorsal striijc, coutractins' toward the horns, becoming narrow behind them, but 

 widening to the tljird abdonnual segment, eontracting to a yellowish point on the hinder edge ot 

 the fourth segment, where a second white band begins in a yellowish point, and is widest on the 

 sixth segment, becoming forked on the last four abdominal segments. The suraual plate is white 

 above; the anal legs white above, lined with red. On the sides of the third thoracic segment i.s 

 a short curved red line. A large, broad oblicpie stripe, lined with red and wiiite, on the side of 

 the fourth abdominal segment. The thoracic legs are stained with pale red; the middle 

 abdominal legs are pale green, stained with red above the plauta'. The sides of the body are 

 tincly dotted with dark blood-red. 



Last stage. — Described August 11. Length, 24 mm. Head high, conical, ending on the vertex 

 in two rounded tubercles, which are blackish; sides of the head reddish brown, the color of a 

 dead maple leaf; extending ni) and down the front is a white median band, including the labrura 

 and clypeus, and extending back over to the white dorsal band beginning on the prothoracic 

 segment. Body of a delicate pea-green, with the characteristic white dorsal band, and on the 

 sides of the third thoracic and first abdominal segments the oblique sere-brown patch inclosing 

 the tirst abdominal spiracle. A second large sere-brown patch on the side of the third abdominal 

 segment, extending forward upon the second, and backward, forming a larger spot on the fouith 

 segment. A third oblique sere-brown patch is situated on the side of the sixth segment, inclosing 

 the spiracle and partly, like those in front, edged irregularly with a darker reddish brown tint. 

 Tlie white dorsal baud is as before, but more pronounced, being wider and more continuously 

 edged with reddish brown. There are fine concolorous dots scattered over the sides of the body. 



A larva which Dr. Dyar thinks may be a variety of H. hiundala was observed on the red oak 

 at Brunswick, Me., August '2,1. Length,. 35-40 mm. The head is not so wide as the body, greenish, 

 with a i»iiik strii)e on each side. The body is thickest in the middle, a double dorsal pinkish 

 brown stripe inclosing a median white line and extending upon the first abdominal segment. 

 There are two subdorsal yellowish strii>es which are interrupted 0!i the third abdominal segment, 

 which is pinkish brown on the sides, but not above, while the succeeding segment is brownish 

 above but not on the sides, being bounded laterally by a yellowish line. On the fifth abdominal 

 segment is a broad dorsal, V-shaped, brown spot, the apex pointed auteriorlj'. Segments 7 and 8 

 with a dorsal brown patch. On the last segment are two narrow brown lines. The sides of the 

 anal legs are rusty reddish brown ; an oblique reddisli brown band extends from the first abdominal 

 upon tlie third thoracic segment and incloses the first abdominal spiracle; the third spiracle is 

 included in an oblicpie brown band extending from the brown dorsal spot above. An oblique 

 lateral brown band on the sixth abdominal segment, and on the seventh and eighth is a lateral 

 oval concolorous spot connecting the two segments. The ground color of the body is pea-green, 

 speckled finely and densely with black. 



The rust red brown spots are of almost exactly the same hue as the rusty sere spots on an 

 oak leaf, and it is thus at first difficult to detect the larva when feeding on the edge of a leaf. 



This larva is, I am quite sure, the final stage of JI. hiundata (PI. XXXII, figs. 2, 2rt), as I have 

 found it during the past season on the red or rock maple. One found August 28 was without the 

 oblique pale brown line on the third tlioracic and first abdominal segments, and also without 

 the brown spot on the side of the sixth abdominal segment.' 



• The following descriptiou is drawn up from a specimen sent me by Dr. Dyar, whicli proves to he H. hiundata: 



Leugtli, 35 unn. The head is not so wide as the body. Width, 4 mm. It narrows above more than usual, so 

 that at the vertex it is scarcely more than one-fourth as wide as below. The sides are angular. The vertex is 

 bilobe<l, ending in two well-marked conical tubercles, which are black at the end, but not pilifcrous. 



The head is whitish, polished greenish, with .a faint purplish tinge. The region in front and on each side of 

 the clypeus and from there to the vertex is polished greenish, without the purplish tint. 



The body in general is entirely free from piliferous warts or humps. There are four minute piliferous warts on 

 the first thoracic segment, which is normal. The body is spindle shaped, tapering toward each end, jiosteriorly to 

 the anal legs, which are small, weak, anil polished faint purplish. The suraual plate is small, rounded behind, 

 smooth, with two dark dorsal stripes. The body is pale green. As the example before me is full-fed and about 

 ready to pupate, it is faintly marked with sere and brown, pale raw sieuna on the sides of the tirst abdominal 

 segment, with a small p.itch, and on the sides of the sixth abdominal segment, while the sides of the third and 

 fourth, the whole of the third, the spot encroaching a little on the second segment, are washed with the same iialo 

 brown or raw sienna tint. (Dr. Dyar suggests that the white dors.al marks must have been lost on account of the 

 approach of jiupation.) 



