934 THE AMERICAN NATURALIST. [Vor. XXXVI. 
If the spine from a primary spiral does not increase in length 
rapidly, but increases in width, the secondary and later spirals, 
instead of developing independent spines, encroach upon the 
primary one and modify it by the production of lateral crenula- 
tions, which, as the spine grows, diverge more and more and 
become more and more pronounced, until the striking multi- 
lobed character of the spine of M. pa/ma-rose@ (Lam.) and other 
highly ornate species is attained. If the principal spine of such 
a species is traced backward through preceding varices (which 
can generally be done, since resorption does not reach up to it), 
it will be found to be less and less complex in earlier and earlier 
varices, and ultimately may perhaps — in a very perfect speci- 
men — be resolved into a single spine. (See Fig. 10.) In the 
growth of the spine itself, from its beginning on the final or 
first varix (counting backward) to its full development, as 
already noted, it progressively increases in complexity. A stage 
somewhat earlier than the adult stage in this spine shows the 
same degree of complexity as the adult. of the representative 
of this spine in the next preceding or second varix. A still 
earlier stage in the principal spine of the first varix corresponds 
to the stage just preceding the adult in the principal spine of 
the second varix, and to the adult stage in the corresponding 
spine of the next earlier or third varix, and so on. Inthe same 
manner, though less perfectly developed in most cases, the 
spine next below (anterior of) the principal spine on the final 
or first varix has the characters of the adult of the principal 
spine in the next preceding or some earlier varix, and the third 
spine has the adult characters of the principal one in a still 
earlier varix. The last spine of the final varix — presumably 
the last formed one — has the characters of the principal spines 
in one of the earliest varices, when the shell was still very 
young (Fig. 10, where 4-7o corresponds to K-z, and -ro to 
L-r). Thus each spine passes through a succession of stages, 
and in its reappearance in a new varix it has made marked 
advances. The life history, therefore, of a single spine may be 
read by noting the characters of all the spines of that varix 
progressively from the smallest to the largest. This also indi- 
cates in general the life history of the group to which the 
