196 THE entomologist's record. 



Chapman's latest paper (already referred to) has given us fresh 

 clues for the real evolutionary stems along which the Lepidoptera 

 have probably been evolved, and if it were not for the uncertainty 

 relating at present to the Geometkides and Platypterygides, a very 

 fairly complete genealogical tree might be constructed, comprising, at 

 least, the principal families. Chapman's paper distinctly suggests 

 three main stirpes, derived from a hypothetical base, which we may 

 term : (1) The Sphingo-Limacodid stirps. (2) The Noctuo-Hepialid 

 stirps. (3) The Pyralo-Micropterygid stirps. It is very doubtful 

 whether the ERiocEPHALmKS do not really represent a very ancestral 

 form of the Pyralo-Micropterygid stem. At present, of course, they 

 are considered as a separate sub-order, by Packard, and the arrange- 

 ment would then fall. 



Sub-order I. — Lepidoptera-laciniata (Proto-lepidoptera). — 

 Eriocephalid stirps. 



Sub-order II. — Lepidoptera-haustellata (Pal.eo- and Neo-lepidop- 

 tera). — 1. Sphingo-Limacodid stirps. 2. Noctuo-Hepialid stirps. 

 3. Pyralo-Micropterygid stirps. As I have just stated, the super- 

 families, about which one feels the least certainty at present, are the 

 Geometrides and PLAiYPXERYGroEs. We had always considered the 

 former somewhat closely allied to the NocTumEs, and the latter to the 

 Pyralides. The egg- structure of the Geometrides forbids the associa- 

 tion of this super-family with the Noctuides, and we are inclined to 

 think it should be connected with the Sphingo-Limacodid stirps. The 

 PLATYPTERY(imEs still appear to us to belong to the Pyralo-Microp- 

 terygid stirps. 



As matters stand at present, it would appear that we are on pretty 

 safe ground in making the following associations : — 



Stirps 1. SpmNGo-LiMACODiD Stirps. — CdcJiliopodides, Xi/;iaeni(h's, 

 LamH-atnpidt.s, (Tcoinctridcs (/), Kudidiiiidcs, Satur- 

 iiiidi's, Sp/ihit/idi's. 

 Stirps 2. Noctuo-Hepialid Stirps. — llcjiialidcs, Zeu^eride-s, Cossides, 

 Tortricides, i^i/ctaeolides, NalideK, Notod(»itides, JJiui- 

 ride.s, Arctiides, Noctuides, Hesperidesi, Papilionides. 

 Stirps 3. Pyralo-Micropterygid Stirps. — Micropteri/i/ides, Ade- 

 lides, Tineides, Scmdct {/), Psi/chides {/), ColeopJiorides, 

 Crambides, Pyralides, Platypterygides ( ! }. 



On the structual characters of Lepidopterous Larvje. 



By HAIUIISON G. DYAE, Ph.D. 



I. Larv.e of Bombyces. — The larvie of theBombyces have tubercles "•••= 

 iv and v remote, iv almost always moved well up behind the 

 spiracle. The group is a large one, and covers a considerable range 

 of characters. The most generalised families have simple single seta}, 

 the snbprimary ones present. This condition is seen in the Noto- 

 dontidfu, Noctuida?, Geometridi?, Agaristidse, etc., which are very 

 uniform, and give no characters which will separate the individual 

 families. Specialisation takes place in the conversion of the simple 

 tubercles into warts with many hairs. This has taken place in two 

 types. In the first there are seldom any secondary hairs, and the warts 



* The positions of these tubercles are described ante, p. 130. 



