Zoea-form in different families, in which the earlier stages of
development are effaced. But except what is common to all Zoeae, and
what may easily be explained as being transferred back from a later into
this stage, the Zoeae of the Crabs, for example, agree with those of
Pagurus and Palaemon in no single peculiarity of structure which leads
us to suppose a common inheritance. Consequently we may apparently
assume, without hesitation, that when the Brachyura and Macrura
separated, the primitive ancestors of each of these groups passed
through a more complete metamorphosis, and that the transition to the
present mode of development belongs to a later period. With regard to
the Brachyura, it may be added that in them this transition occurred
only a little later and indeed before the existing families separated.
The arrangement of the processes of the carapace, and, still more, the
similar number of the caudal setae in the most different Zoeae of Crabs
(Figures 19 to 23) prove this. Such an accordance in the number of
organs apparently so unimportant is only explicable by common
inheritance. We may predict with certainty that amongst the Brachyura no
species will occur which, like Peneus, still produces Nauplius-brood.*
(* I must not omit remarking that what has been said as to the
development of the Crabs applies essentially only to the groups
Cyclometopa, Catometopa and Oxyrhyncha, placed together by Alph.
Milne-Edwards as "Eustomes." Among the Oxystomata, as also among the
"Anomura apterura," Edw., which approach so nearly to the Crabs, I am
unacquainted with the earliest young states of any of the species.)

As we have already seen, Mysis and the Isopoda depart from all other
Crustacea very remarkably by the fact that their embryos are curved
upwards, instead of, as elsewhere, downwards. Does not so isolated a
phenomenon as this, it might be asked, in the sense of Darwin's theory,
indicate a common inheritance? Does it not necessitate that we should