—The Line of Descent of the Invertebrata. 283 
— Protozoa.—In this group are included the Foraminifera and 
Radiolaria. As regards the former, the author points out the com- 
parative scarcity of their remains in the older Paleozoic rocks and 
their comparatively sudden appearance in great variety in Car- 
boniferous strata, in which some highly organized forms, as Num- 
mulites, Fusulina, etc., are associated with Modosaria, Textularia and 
Astrorhizida, etc. The forms with calcareous shells are all regarded 
as derived from those with agglutinated tests, and these again have 
probably sprung from types with extremely simple and irregular 
tests like those of the Astrorhizidee. 
In the brief notice given of the Radiolaria, it is mentioned that 
the forms recently discovered in Jurassic and Triassic strata show 
but small differences from those of Tertiary and recent seas, and 
_ afford no clue to the natural relationships of the various divisions. 
In the Ccelenterate division are embraced the Sponges, Corals, 
Hydrozoa, and Graptolites. As regards the Sponges, the author 
adopts with certain modifications the views of Zittel, and concludes 
with diffidence that the siliceous forms are probably derived from 
some more ancient type, in which the elementary spicule of the 
skeleton was irregularly polyaxile. 
The Corals are treated very fully, and due importance is given to 
the fresh light thrown upon their characters by the microscopic 
observations of recent students. The Rugosa or Tetracorallia are 
regarded as a sharply defined group characterized by the symmetri- 
cally bilateral arrangement of their septa, and they are believed to 
be restricted to Paleozoic strata. Attempts are made to show that 
the forms from Mesozoic rocks which have been relegated to this 
group do not really belong to it, and conversely that Palzozoic 
Corals which have been placed with the Hexacorallia should really 
be included with the Rugosa. Some of the statements respecting 
the sharp division between the Rugosa and the so-termed Hexa- 
corallia, namely, that the young forms of the former group are 
always symmetrically bilateral, and that in the latter the septal 
skeleton is purely radial and at the commencement with six septa, 
are too comprehensive and can hardly be substantiated. The genus 
Calostylis from Silurian strata, which, by Lindstrom, Nicholson and 
other excellent observers, is considered to be an undoubted member 
of the Perforata, is cast out of this division by Neumayr on the 
suspicion that it possesses bilateral symmetry, and that it is thus a 
Perforate Rugose Coral; but it is probable that the objection to 
ranking it with the normal Perforate Corals mainly depends on its 
occurrence in the Paleozoic rocks. The Palaeozoic Favositoid Corals 
are regarded as forming an independent group, on the ground that 
the perforations in their walls are morphologically entirely different 
from those in normal Perforate Corals, and we think the objection 
a good one. The genera Heliolites and Halysites are, on the grounds 
of their possessing twelve true septa, indicating a corresponding 
number of true mesenteries and tentacles, considered as nearly 
related to the Fistuliporide and without any real affinity to the 
Alcyonarian genus Heliopora, with which they have been associated 
