152 Inheritance in Philosamiu Ili/bn'ds 



ncit unreasonable to suppose that all the more specialised genera must 

 have been derived from a Saturnia-like ancestral form, i.e. a larva in 

 which all the tubercles, whether the thoracic or the abdominal, were of 

 the same shape, size and colour on all the segments; those on the 

 eighth abdominal segment being of the same number (6) as on the 

 segments in front"; p. 19, "The larva oi Aglia after its last moult 

 becomes like a Saturnian, possibly of the type of Telea and especially 

 like a smooth form of the larva of Attacus betis (Walker) (also called 

 Rothschildia). It is clear that the Saturnian larva, although well 

 tubercled in the earlier stages, has, as a whole, a less complicated tubercle 

 in the last stage than the Ceratocampidae and that it is only in the 

 Saturniidae that a number of forms occur which have much reduced 

 tubercles in the last stage." 



The complete reduction of tubercles (excej)t on one segment) has 

 reached its limit in two forms, one belonging to the Ceratocampidae, 

 viz. Cerctrphana frauertfeldii {venasta variety) and one of the Satur- 

 niidae, viz. Rhodia fugax. In appearance these two are remarkably 

 alike and so abnormal in appearance that they must be regarded as 

 highly specialised forms which have approached one another perhaps 

 because of a similar environment. These forms being so highly 

 specialised it is necessary to seek the relation of the present non- 

 tubercled hybrid, if any relation exists, among the more ordinary forms 

 which occur in the Saturniae in which the tubercles are much reduced 

 rather than entirely absent, such as Telea liulyphemea, Callosamia 

 ■promethea, Tropea luna, Aglia tau, etc. 



Apparently a form with simplified tubercles, similar to Aglia, arose 

 as a mutation within the Ceratocampidae, and from this form some of 

 the vSaturniidae were certainly descended — it is difficult when a study is 

 made of the larvae, t(3 suppose that all the Saturnians are derived from 

 the same ancestor since they are divided into two natural gi-oujjs by 

 the condition of the tubercles on the eighth abdominal segment. 



Has there, been a repetition de novo in this hybrid of the same 

 phenomenon which produced Aglia tail,, or through the ages since the 

 latter originated, have the gens for a reduced tubercular condition in 

 the last instar, been waiting in ricini or in cyathia ? 



My thanks are due to the Natural History Society of Birmingham 

 and to the British Association for grants in aid of this research. 



