THEORY OP THE EMBRYONIC PHASE OF ONTOGENY. 331 



always large and biramous, whilst at the same time some of the 

 most posterior segments have been modified to form a powerful 

 jointed " tail." The thorax retained its primitive character, 

 and is accordingly suppressed in the larva; though here in 

 comparing the Zosese of the various groups we meet with a 

 precise parallel to the case of the Nauplii. The Zosea of the un- 

 differentiated Schizopod possesses only one pair of maxillipedes, 

 and even they are short and somewhat foliaceous, but it shows a 

 distinct segmentation of both thorax and abdomen. Still more 

 instructive are the larvae of the lower families of the Decapods, 

 the Sergestidse and Penseidse. Taking Penseus for example, we 

 find that it escapes from the egg-membrane as a Nauplius : it 

 gradually changes to a Zoaea with two pairs of maxillipedes and 

 the thorax distinctly segmented and with rudimentary append- 

 ages ; this passes into a form with thorax well developed and 

 all its appendages biramous — the so-called Mysis-stage, closely 

 resembling the adult Schizopod, — and from this it passes to the 

 adult state. On the other hand, in the highly specialised 

 Brachyura we find a highly specialised Zosea, in which the 

 thoracic segments are totally suppressed and the thorax pro- 

 longed into great spines, the Mysis-stage is dropped but a new 

 " Megalopa "-stage is introduced, which strongly recalls the 

 Macrura, and may be taken to indicate a Macrurous ancestor. 

 The existence of the Megalopa and Mysis stages, the signifi- 

 cance of which is obvious, affords the strongest reason for 

 maintainingthe ancestral significance of the Nauplius and Zosea 

 stages; in doing so one merely follows the universal rule of 

 science, i. e. reasoning from the known to the unknown. 



Turning now to the embryonic type of development, let us 

 examine the causes which are likely to modify a course of 

 development which is primitively larval. First we must 

 discriminate between various kinds of embryonic development. 

 There is, in the first place, the type in which the organism is 

 confined within the egg-membrane and supplied with nutriment 

 by means of yolk stored up in its cells. Secondly, we have 

 cases in which the embryo, still remaining in the egg-mem- 

 brane, is retained in the body of the mother, the egg being 



