86 PROF. W. GARSTANG ON THE THEORY OF RECAPITULATION : 
8. That " recapitulation " does not require the reproduction of adult stages 
n the ontogeny in order to be exhibited is plainly seen in the development 
of many Georaetrid moths. Everyone knows the Greometrid or " looper " 
type of caterpillar, provided with prologs only on two of the hindmost 
abdominal segments (the 6th and last). This type is admittedly derived 
from a prototype which possessed the full Lepidopteran equipment of prolegs 
on segments 3 to 6, as well as the last, the prolegs on the three first segments 
having subsequently disappeared. But many Greometrid caterpillars possess 
vestiges of one or more pairs of these missing prolegs : in the March moth * 
(^Erannis ci'scularia, SchifE.) there are traces of the last pair (South, ii. pi. 125); 
in the common Brimstone ( Opisthogra'ptis cratcvgata (Linn.)) and Scalloped 
Hazel (Gonodontis bidentata, CI.), clear rudiments of the last two pairs {I. c. 
pi. 115) ; while the Orange Underwing [Bi^eplios parthenias (Linn.)) has the 
first two pairs rudimentary and the third pair fully developed and functional 
(Meyrick, 1895, and South, 1908, ii. pi. 39). In the Feathered Thorn, Colotois 
(Himera) pennaria (Linn.), the single pair of vestigial prolegs arises and 
disappears between the 1st and 4th moults (Buckler, 1897). Now, the time 
has long passed when it was possible to regard these prolegs as homologues 
and derivatives of the true legs of some Scolopendroid ancestor. They were 
" cenogenetic " larval features, adaptive interpolations, modifications of the 
middle stages of a life-cycle which originally, in the earlier phases of Endo- 
pterygote history, exhibited larvae lacking prolegs altogether, as in Coleoptera. 
Yet these examples of vestigial organs are as reminiscent of ancestral (though 
larval) structure as the larval foot of the oyster, the larval stalk of Antedon, 
the transitory feet of the parasitic Portunion, or any other of the familiar 
examples that are held to prove the theory of adult recapitulation. They 
demonstrate, as Morgan has already urged (1919), that recapitulation is 
merely the static aspect of inheritance, and that, in this aspect, inheritance 
is not primarily the reproduction of adult characters, but the reproduction 
-of the characters of each part of the whole life-cycle — the sequential 
expression of the full train of zygotic potencies. 
9. It may be urged that such hereditary changes in the middle phases of 
the life-cycle do not affect the proposition that evolutionary changes usually 
take place at the end, and that the case for adult recapitulation rests on the 
evidence for this proposition. Nevertheless, to clear up the misunder- 
standings of the past, it is necessary to leave no margin for ambiguit3\ 
If " recapitulation," in the special sense of partial reproduction of the past, 
is hereby shown to be independent of the characters of adult ancestry, that is 
something gained : the axe is laid at the root of the tree. For much of the 
glamour of the old biogenetic law is due to its appeal to such idols of the 
market-place as the assumj^tion that " like begets like," and that, as adults 
* Meyrick's classification aud nomenclature are followed here, but the English names 
have been added for convenience of reference to South's fia'ures. 
