221 
berg by the closely allied S. radiata, Vanuxem, and S. Wool- 
worthana, Hall. 
Evidence of precisely similar import can be obtained from a 
consideration of the distribution of the species of the genus 
Spirifera within the same formations. The Spirifera crispa, 
Linn., of the Niagara formation, is succeeded in the Lower 
Helderberg by the closely allied, if not identical, S. cycloptera 
and S. Vanuxemi, Hall, which are followed in the Oriskany by 
S. tribulis, Hall, and in the Corniferous by S. duodenaria. Hall. 
Again, the Spirifera sulcata of the Niagara group is succeeded 
by the cognate S. perlamellosa, Hall, of the Lower Helderberg, 
in turn followed in the Corniferous group by S. raricosta, 
Conrad. Lastly, the Spirifera Niagarensis, Conrad, which, so far 
as I am aware, is the oldest example of the genus in the Silurian 
rocks of North America, is directlv succeeded in the Lower 
Helderberg by the closely related S. macropleura, Conrad. 
The question now arises — What is the significance of facts 
such as these, — facts which could be greatly multiplied, and 
which no competent authority would think of disputing? Are 
we to consider that the eighteen forms which group themselves 
round Orthis hybrida as a central type, and which are found in 
the successive formations from the Clinton to the Hamilton, 
are so many absolutely distinct species, in the old and strict 
acceptation of this term ? Or, shall we simply expand our 
conceptions of what constitutes a species, extend the limits of 
the term, and consider that these allied forms are so many 
more or less distinct varieties of a single protean species? If 
the latter view were adopted, whilst to the working palaeonto- 
logist these forms would remain as so many distinct species, 
and would properly and usefully be designated by so many 
distinct names, to the transcendental palaeontologist they would 
become simply so many successive phases of one variable form. 
It cannot be too strongly borne in mind, as very properly 
insisted on by Mr. Darwin, that naturalists have no golden rule 
for determining what are species and what are varieties. Such 
determinations rest upon the value which certain observers 
attach to certain characters ; and this is especially true of 
fossils, where, in addition to the actual anatomical or structural 
characters, we have the additional element of time introduced. 
Specimens which would at once be admitted to be mere varieties 
if they occurred in a single stratum, are by many palaeonto- 
logists unhesitatingly set down as distinct species, if they 
happen to occur in beds of even slightly different age. For 
the purposes of the stratigraphical geologist, this does not 
matter, and is, indeed, often useful, since if a certain bed can be 
VOL. ix. n 
