TRANSACTIONS OF SECTION D. 823 
ancestral forms comparatively unchanged. Workirg upon this material, com- 
parative anatomy and embryology can reconstruct for us the general aspects of a 
history which took place long before the Cambrian rocks were deposited. This 
line of reasoning may appear very speculative and unsound, and it may easily 
become so when pressed too far. But applied with due caution and reserve, it 
may be trusted to supply us with an immense amount of valuable information 
which cannot be obtained in any other way. Furthermore, it is capable of stand- 
ing the very true and searching test supplied by the verification of predictions 
-made on its authority. Many facts taken together lead the zoologist to be- 
lieve that A was descended from C through B; but if this be true, B should 
possess certain characters which are not known to belong to it. Under the in- 
spiration of hypothesis a more searching investigation is made, and the characters 
are found. Again, that relatively small amount of the whole scheme of animal 
evolution which is contained in the fossiliferous rocks has furnished abundant 
confirmation of the validity of the zoologist’s method. The comparative anatomy 
of the higher vertebrate classes leads the zoologist to believe that the toothless 
beak and the fused caudal vertebrae of a bird were not ancestral characters, but 
were at some time derived from a condition more conformable to the general plan 
of vertebrate construction, and especially to that of reptiles. Numerous secondary 
fossils prove to us that the birds of that time possessed teeth and separate caudal 
vertebre, culminating in the long lizard-like tail of Archzeopteryx. 
Prediction and confirmation of this kind, both zoological and paleontological, 
haye been going on ever since the historic point of view was adopted by the 
naturalist as the outcome of Darwin's teaching, and the zoologist may safely claim 
that his method, confirmed by palzontology so far as evidence is available, may be 
extended beyond the period in which such evidence is to be found. 
And now our last endeavour must be to obtain some conception of the amount 
of evolution which has taken place within the higher Phyla of the animal kingdom 
during the period in which the fossiliferous rocks were deposited. The evidence 
must necessarily be considered very briefly, and we shall be compelled to omit the 
Vertebrata altogether. 
The Phylum Appendiculata is divided by Lankester into three branches, the 
first containing the Rotifera, the second the Chzetopoda, the third the Arthropoda. 
Of these the second is the oldest, and gave rise to the other two, or at any rate to 
the Arthropoda, with which we are alone concerned, inasmuch as the fossil records 
of the others are insufficient. The Arthropoda contain seven classes, divided into 
two grades, according to the presence or absence of antennze—the Ceratophora, 
containing the Peripatoidea, the Myriapoda, and the Hexapoda (or insects); the 
Acerata, containing the Crustacea, Arachnida, and two other classes (the Pantopoda 
and Tardigrada) which we need not consider. The first class of the antenna- 
bearing group contains the single genus Peripatus—one of the most interesting 
and ancestral of animals, as proved by its structure and development, and by its 
immense geographical range. Ever since the researches of Moseley and Balfour, 
extended more recently by those of Sedgwick, it has been recognised as one of the 
most beautiful of the connecting links to be found amongst animals, uniting the 
antenna-bearing Arthropods, of which it is the oldest member, with the Cheetopods. 
Peripatus is a magnificent example of the far-reaching conclusions of zoology, and 
of its superiority to paleontology as a guide in unravelling the tangled history of 
animal evolution. Peripatus is alive to-day, and can be studied in all the details 
of its structure and development; it is infinitely more ancestral, and tells of a far 
more remote past than any fossil Arthropod, although such fossils are well known 
in all the older of the Paleozoic rocks, And yet Peripatus is not known as a 
fossil. Peripatus has come down, with but little change, from a time, on a mode- 
_ Tate estimate, at least twice as remote, and probably many times as remote, as 
the earliest known Cambrian fossil. The agencies’ which, it is believed, have 
crushed and heated the Archzean rocks so as to obliterate the traces of life which 
they contained were powerless to efface this ancient type; for, although the passing 
generations may have escaped record, the likeness of each was stamped on that 
