smell of suutf. These scent-brushes at the end of the finger-shaped pro- 

 cesses are similarly found in Lycorea, and also (although less developed and 

 smelling weaker) in Daiiais GU'qrpus and D. Erlp2)us, in which species they 

 appear to have been hitlierto overlooked. 



" Thus, on the ground of the foregoing characteristic differences, and 

 especially of the former, which, being obviously of no advantage to the 

 species, may be takeu as sure indications of a coimnon origin, the Ithomiae 

 and the true Danaidpe (Danais, Lycorea and Ituila — Hestia and Euplaa, 

 1 know only from figures) must be regarded as two groups which long 

 ago underwent separation, and which are at least as far apart as, perhaps, 

 the Acraeiuse and the Maracuja butterflies. These last groups are also 

 distinguished by the wing-cell between the veins \h and 2 of the hind- 

 wing, which is double in the Acrpeinae as in the Ithomine, and single in the 

 Maracuja group as in the Danaidae. The caterpillars of Acraeinae accord 

 completely in character with those of the Maracuja butterflies, but not with 

 those of the DanauUie and Ithomiae ; the former so well known as living on 

 Asclepiadaceae have on the back two (Danais Erippus), three (D. Gilippus), 

 or four ( Euplma Midamus) pairs of long, thread-like, unprotrusible ' tentacles.' 

 The Ithomia caterpillars, which feed on Solanaceae or the nearly-allied 

 Scrophulariaceae, are either entirely without appendages or have below the 

 spiracles fleshy spherical protuberances {Mechanitis Lysbnnia)S- 



" If, in accordance with all characters, Thyridia is thus related to 

 Ithomia, and Ituna to the Danaides (if the latter is not, as in Kirby's 

 ' Catalogue,' separated from Lycorea by the Ithomiau genus AtJiesis), the 

 resemblance of these two genera could only be ascribed to descent from a 

 common ancestor if they had preserved the original colouring and marking 

 of the ancestors of all the Ithomiae and Danaidae. But this cannot be 

 maintained. If the progenitors of these two groups had possessed wings 

 witb large transparent spaces, it is improbable that such a large number of 

 the existing species of the said groups should have reverted to a still earlier 

 type of wing completely clothed with scales. It might be maintained, with 

 equal right, that Lycorea and the various Ithomiae, so similar to this genus 

 in marking and colouring, indicate the original marking and colouring of 

 the groups in question. 



"A case 6i acquired resemblance — one of imitation or mimicry — is thus 

 presented. But which of the two species, Ituna Ilione or Thyridia Meyisto, 

 is the original form, and which the mimic? On this point there should 

 surely be no doubt. Does not a species which serves as a model occur 



* The caterpillar represented by Boisduval (Spec. Gen. Lepidopt., pi. 4, fig. 9), 

 ascribed to Stalachtis [Nerias) Kuterpe, appears to be that of a Mechanitis ; in 

 hairiness it resembles that of M. Lysimnia. A glance at this figure and figs. 10 

 and 11 of the same plate, which represent caterpillars of true Danaidae, shows 

 immediately the great difference between the caterpillars of the Danaidae and 

 Ithomiit;. 



