EEPORT ON THE TUNIC AT A. 
29 
describes and figures their separate and independent existence in Diazona arid Distoma, 
two of the genera of his Tethyes Composees. Clavelina in his system is placed— rightly, 
I consider — next to the “ Phallusise Cionse ” (the modern genus dona) among the Ascidise 
Simplices. 
Savigny classified the ten genera which he recognised amongst Compound Ascidians 
by means of characters taken from the branchial and atrial apertures. But although 
such characters are most useful and constant marks of affinity in the Simple Ascidians, 
they fail signally as applied by Savigny to the Compound forms, and result in the 
separation of his closely allied genera Didemnum and Eucoelium, while Diazona, 
Distoma, and Sigillina are thrown together in one group, although really belonging to 
distinct families, and Euccelium is placed with Botryllus, a genus with which it has 
certainly no close relationship. 
Lamarck’s classification of the Tunicata, pubhshed about the same time, was based upon 
the arrangement of the Ascidiozooids in the colony, but the result was no better than 
that obtained by Sa\dgny’s method, since Pohjclinum was united with Polycyclus and 
Botryllus, Euccelium was united with Aplidium, and Distomus was grouped along with 
Sigillina, while Aplidium, Polyclinum, and Sigillina were widely separated. 
Cuvier, writing shortly afterwards, refused to accept the majority of Savigny’s and 
Lamarck’s genera, on the ground that they were not sufficiently distinct from one another. 
Subsequent investigators have not supported him in this view. Savigny’s genera are 
still nearly all retained, and some have even been broken up into several groups now 
regarded as distinct genera. 
The next classification of importance is that of Milne-Edwards, published in 1841, 
and one of the most notable features of his arrangement is that it involves the separation 
of the Clavelinidse (at that time the two genera Perophora and Clcwelina) as a distinct 
group, the Ascidim Sociales, occupying an independent position between the Simple and 
the Compound Ascidians. He defined this new group as comprising Ascidians which 
reproduce by buds as well as by eggs, and which live united by common radiciform 
prolongations, but which otherwise are free of all adhesion to one another. He distin- 
guished the Simple Ascidians as forms which never reproduce by gemmation and are 
never found in groups united by a common tegumentary tissue ; while he separated the 
Compound from the Social Ascidians on account of their possessing a test common to all 
the members of the colony. If we unite the Simple and Social Ascidians, which, as I have 
shown in the first part of this Eeport,’- there is reason for doing, we shall have, according 
to Milne-Edwards, the Simple and Compound Ascidians distinguished merely by the 
members of the colony in the latter being united by a common test, while in the former 
each individual has its own distinct tunic. This character, although much better than the 
one made use of by Savigny, is, as we shall see later on, by no means an infidlible guide. 
1 See vol. vi. part xvii., where the Clavelinidje are treated as a family of the Ascidise Simplices. 
