1897] SOME NEW BOOKS 127 
the septa and the coenenchyme. For her the form and structure of 
the calcareous laminae or bars are the result of foldings, wrinklings, 
and tuckings in and over of the flexible zooid. On p. 315 there is an 
imaginative description of the evolution of the madreporarian zooid, 
which is represented as pulling in and tucking up its body in various 
places, forming invaginations here, evaginations there, as if guided by 
some predetermined impulse, and we are to believe that the form of 
the corallum is determined by these almost purposive wrinklings and 
corrugations of the zooid, which fills up the cavities and folds of the 
creases in its body with calcareous tissue. References, scattered 
throughout the volume, to invaginations—a word, by the way, 
which is used in a most puzzling variety of meanings with regard 
to spatial relations—show that Miss Ogilvie is dominated by the 
idea that the wrinkling and pitting of the soft tissues was the 
antecedent, the formation of calcareous structures the consequent. 
The skeleton, it is true, is formed by the soft tissues, but it by no 
means follows that the form of the skeletal parts is the result of the 
pre-existing form of the soft parts. The two elements have been 
formed part passu, changes in the one reacting upon the other, and 
the final shape and mutual relations are the result of a continuous 
correlated development of which we cannot affirm, at any given 
point, that the growth of the one part preceded or dominated the 
growth of the other. On Miss Ogilvie’s supposition it is most 
difficult to account for the formation of the canal system in perforate 
corals, and we cannot but suspect that this difficulty has led to her 
giving a theoretical rather than an actual picture of the structure of 
Turbinaria and of Fungia. Not that Fungia is a perforate coral. 
In classifying the Fungidae, Miss Ogilvie has left out of con- 
sideration the fact that the young Fungia is a true aporose coral, 
indistinguishable from a Turbinolid, even to the absence or at least 
the very slight development of synapticula. This fact points to a 
close relationship between the Fungidae and the Turbinolidae, yet 
they are classed far apart, the former among the Pollaplophracta 
Septocorallia, the latter among the Haplophracta Murocorallia. In 
fact, the more one examines the grounds of the classification adopted 
in this work the less satisfactory does it appear. The group 
Coenenchymata strikes one as purely artificial. The subsections 
Murocorallia and Septocorallia are founded on the presence or 
absence of so-called ‘thecal’ pieces. This is a partial revival of 
the classification proposed by von Heider and adopted by Ortmann, 
it has been severely criticised by von Koch and others, and it is not 
too much to say that it is founded on a misconception. There is 
no essential difference between ‘thecal’ and ‘septal’ structures. 
Both are formed in the same manner from the same regions of the 
polyp. In many forms sections taken low down in the corallum 
show an alternation of septal and apparently thecal pieces. Higher 
up it is found that calcareous lamellae project inwards from the 
supposed thecal pieces, so that the last named appear as septa. The 
same coral appears to have an ‘eutheca’ in one part, a ‘pseudotheca ’ 
in another part, is therefore a Murocorallian ‘in one region, a Septo- 
corallian in another. This fact has been repeatedly emphasised by 
von Koch, and a clear and convincing discussion of the question is to 
