SUPPLEMENT. 
499 
however, appear probable that in a family of fossils so closely allied as 
are all the proper Graptolitidece , any such great diversity in mode of growth 
would exist. 
It will appear evident from what follows, that heretofore we have been 
compelled to content ourselves, for the most part, with describing frag* 
ments of a fossil body, without knowing the original form or condition 
of the animal when living. Under such circumstances, it is not surprising 
that various opinions have been entertained, depending in a great measure 
upon the state of preservation of the fossils examined. The diminution in 
the dimensions, or perhaps we should rather say in the development, of 
the cellules or serrations of the axis towards the base, has given rise to 
the opinion advanced by Barrande, that the extension of the axis by 
growth was in that direction, and that these smaller cells were really in 
a state of increase and development. In opposition to this argument, we 
could before have advanced the evidence furnished by G. bicornis, 'G„ 
ramosus, G. sextans , G. furcatus, G. tenuis , and others, which show that 
the stipes could not have increased in that direction. It is true that none 
of the species figured by Barrande indicate insuperable objections to this 
view ; though in the figures of G. serra ( Brongniart), as given by Geinitz, 
the improbability of such a mode of growth is clearly shown. 
It is not a little remarkable that with such additions to the number of 
species as have been made by Barrande, M‘Coy and Geinitz, so few ra* 
mose forms have been discovered; and none, so far as the writer is aware, 
approaching in the perfection of this character to the American species. 
Maintaining as we do the above view of the subject, which is borne 
out by well-preserved specimens of several species, we cannot admit the 
proposed separation of the Graptolites into the genera Monograpsus, Di- 
plograpsus and Cladograpsus, for the reason that one and the same species* 
as shown in single individuals, may be monoprionidean or diprionidean, or 
both; and we shall see still farther objections to this division, as we 
progress, in the utter impossibility of distinguishing these characteristics 
under certain circumstances. We do not yet perceive sufficient reason to 
separate the branching forms from those supposed to be not branched; 
