270 



THE ENTOMOLOGIST. 



The meso- and post-thoracic tubercles approximate in number, 

 differing in arrangement from the abdominal tubercles, are not 

 usually treated as homologous with them. This probably is due 

 to the earlier opinion of Dr. Dyar, who did not consider them to 

 be so, and gave a different set of identification numbers. Later 

 authors, using Dyar's work as a foundation, repeat the formula. 

 Apparently this has been done in the ' Catalogue Lep. Phalcense,' 

 vol. i., thus : meso- and post-thorax adult characters, " a sub- 

 dorsal tubercle i with two hairs (i a + i b) ; ii with two hairs 

 often separate (Ha + iiB); iii, iv, v in line iv and v conjoined; 

 vi subspiracular with one or two hairs. ... In first stage 

 . . . tubercles iii and iv of thorax and vi of abdomen absent." 



Dr. Dyar now admits * the thoracic and abdominal tubercles 

 are homologous, but because it is fourth in sequence makes ii b 

 of thorax =^ iv of abdomen. I submit the homologue of iiB of 

 thorax is a minute anterior supraspiracular tubercle of abdomen, 

 called by me iii b; that Dyar's iii of thorax = a subspiracular 

 tubercle of abdomen ; and so on. As stated by my friend Mr. 

 Bacot {ante, p. 94), my iii b " is of very general, if not universal, 

 occurrence on the abdominal segments of lepidopterous larvae " ; 

 but I persist in my opinion, and after examination of more 

 material in the first larval stage, the conclusion is more irresist- 

 ible that it is correct. 



When the system of numbering the abdominal tubercles 

 originated, the minute tubercle (iiiB) was overlooked, or con- 

 sidered of no importance ; but in Melanchra, second stage (figs. B 

 and 4), the positions, by actual measurement, of the fourth 

 thoracic (ii b) and my iii b are identical in longitudinal line. 

 Furthermore, ii b, equally with iii b, is invariably above the 

 spiracular line (figs. 5, 6, 7, 8, 10, 12, 13, 14). Indeed, the 

 positions of larval tubercles are not absolutely fixed in the 

 Lepidoptera, the characteristic dorsal trapezoidal pattern of the 



. .ii 



abdomen, formed by i and ii thus |: , is subject to modifi- 



•ii i- 



cation ; and in Psychidse the pattern is reversed, thus 



.11 



. 'ii' 

 1* 



Hepialidse (fig. 5) first larval stage has the trapezoidal pattern 

 on meso- and post-thorax; my notes give PteropJiorusQ,ho. But 

 the corresponding tubercles of the thoracic segments are usually 

 one above the other, but in longitudinal line with i and ii 

 respectively of abdomen — Cossidae >fig. 7), Pieridse, Tineidae, &c. 

 In Hepialidae, tubercle iiiB has a longer seta than usual 

 (fig. 6a), and in later stages the tubercle is normal in size and 

 seta. The various positions of iii b may here be pointed out, as 

 shown in Hepialidfe (fig. 6a), Tineidse (fig. 9), and Khopalocera 

 (fig. 11). iii B is always minute, so far as my observations go, 

 and often extremely difficult to detect (except in Hepialidse). 



* 'Entomologists' Record,' vol. xiii. p. 40, 



