96 SECTIONAL ADDRESSES. 
simplification in branching follows whereby the stipes are reduced to 
two, and an entirely new factor supervening in localisation of thicken- 
ing in the graptolite wall, the line diverges in one direction and leads into 
the Retiolitide, a quite distinct species-group. 
(b) follows the tendency to simplification, and passes into the two- 
stiped form Didymog. eatensus, and thence to an unsuccessful one- 
stiped graptolite Azygog. eivionicus. The two-stiped form undergoes 
also various modifications in the packing of the cells, and passes 
through Didymog. nitidus, a very variable graptolite, into D. hirundo, 
a more stable form, which is, however, apparently a dead end. In 
others of this Didymog. extensus type an actual modification of the cell 
structure supervenes as an entirely new factor, so that the cell, instead 
of being a simple tube, is gradually bent and twisted and its aperture 
turned in. The details of this have yet to be worked out completely, 
but the general plan is perfectly clear, and leads first into the Leptograpiz, 
and thence into the Dicellograpti, Dicranograpti, and Climacograpti 
in turn. 
Lastly, we may study the elaboration of the cell as seen in the 
second episode, the evolution of the hooked variant of the apertural 
lobe. Here, starting from Monog. cyphus, which has the old-fashioned 
Dichograptus type of cell, we find the first traces of a hook in the 
closely related M. revolutus, and can trace its gradual development in 
the proximal region of the rhabdosoma in M. difformis and M. argenteus, 
in which, though the hook is well developed proximally, the distal 
thece are still simple; gradually the hook-form invades the whole 
rhabdosoma (M. clingani and M, sedgwicki), and taking on its most 
distinctive features in M. marri, reaches its acme in M. priodon, per- 
haps one of the best-known graptolites all over the world. Thereafter 
retrogression sets in, the hook becomes less pronounced in M. flemingii 
s.s., and a small, highly characteristic variety is seen occurring side 
by side with the larger form; this smaller variety gradually gives way 
to Monog. colonus, in which only the proximal thece retain-any signs 
of their former elaboration, and M. colonus itself is replaced by Monog. 
tumescens, where all thece are once more of the unhooked type just 
as in Monog. cyphus, though the form of the rhabdosoma of M. colonus 
is retained. This is one of the latest graptolites with which we are 
acquainted. 
Shallow-water Faunas of the Lower Paleozoic. 
The case of the shallow-water faunas of Lower Paleozoic age must 
now be considered; and here, in spite of a vast amount of work that 
has already been accomplished, much remains to be done, but from a 
different standpoint and along very different lines. There exists already 
a great mass of more or less purely descriptive literature, accompanied 
in general by illustrations of varying degrees of merit. All this has a 
value of its own; it provides descriptions which aid identification of 
fossils, and in many cases gives an excellent idea of the variety of the 
brachiopods, trilobites, or corals represented in a certain bed or set of 
beds; but, looked at broadly, is not its value to a great extent purely 
numerical, giving an idea mainly of the relative abundance of certain 
