their Structure and Affinities. 69 



the contents of the two. This, too, is, as we have seen, precisely the 

 structure of the graptolite ; and if there were nothing more to be 

 taken into account, the fossil could not be separated from the Sertu- 

 lariad . But there is an important additional structure always present 

 in the true graptolite — namely, the slender axis ; and there is no 

 analogous structure in the Sertulariadce. We cannot well compare 

 this axis with that of the FennatuHdce, even neglecting the nature 

 of its composition, unless we make the whole so-called polypary of 

 the Graptolitidid an endo-skeleton instead of an exo-skeleton : but this 

 would introduce a yet more anomalous structure, for which it would 

 not be possible to find any modern ally. 



The comparison, then, that we have instituted between the parts 

 and composition of the polypary of the graptolite and similar poly- 

 paries of living organisms, seems to me to establish beyond doubt 

 that this interesting Silurian fossil has its nearest allies in the Sertu- 

 lariadcc, from which it differs in having a solid axis to the polypary. 



The complete organism presents some characters that at first sight 

 appear to differ from those of the Sertulariadce ; but these difi'erences 

 are not so gi'eat as they are sometimes represented. The genus Den- 

 drograptus has hitherto been found only in fragments in Britain, but 

 the perfect specimens figured and described by Hall have dendroid 

 polyparies with a strong hydrocaulus terminating in a hydrorhiza. 

 This agrees exactly with the habit of Halecium. The other genera 

 of graptolites are generally described as free, but it is remarkable 

 that all of them are furnished with a longer or shorter non-celluli- 

 ferous portion (radicle of Hall) at their proximal termination. The 

 discovery of more perfect specimens shows that this is very generally 

 present, and that in those specimens in which it had been observed, 

 it is longer than has hitherto been supposed. I figure a beautiful 

 specimen of Climacograptus scalaris, Hall (PI. V., Fig. 9), in which 

 this is produced to a considerable length, and there is no indication 

 that even here we have the termination. The relation of this process 

 to the organism shows that whatever may have been its functions, it is 

 homologous with the hydrorhiza ; there is, however, really no reason 

 for supposing that it was functionally different. Hall thinks that the 

 graptolites were free-floating organisms ; Nicholson, adopting this, 

 as he does many other things, and, as his practice is, without acknow- 

 ledgment, is not trammelled by the cautious language of Hall, but 

 asserts that " there can he no doubt that the greater number were free- 

 floating or free-8wimming organisms " (p. 59). And then he finds in 

 the pneumatocyst of the Fhysophoridce the '' best homologue " {sic^ of 

 the central corneous disc of Bichograpsus, and he furnishes the other 

 genera with "possible" swimming bells, which, " o/ course, could 

 never be preserved in a fossil condition " ! Finally, to complete the 

 history, "it must suffice to state, that in the simpler genera the 

 secondary cellules appear to be intercalated between the initial point 



or radicle and the primordial cellule or cellules This mode 



corresponds with that in the .... Physophoridcs''' (p. 58). Un- 

 fortunately, however, it only exists in the author's imagination ; • it 



^ We suppose Dr. Nicholson will allow this, for in a sentence a little before that 



