EEPORT ON THE THNICATA. 
397 
centrally placed common cloacal cavity. The test became penetrated by a well-developed 
system of blood-vessels with enlarged terminal bulbs in the superficial la 3 ^er of the colony, 
forming in aU probability an accessory organ of respiration. This system is evidently 
the same as that found in the test of some of the Ascidiidm, and it has been inherited 
by the BotryUidse from their ancestors amongst the Ascidise Simplices. The Simple 
Ascidians in their turn inherited the blood-vessels of the test from their ancestors the 
primitive Clavelinidae (see fig. 11, p. 388), in which these structures were limited to the 
posterior end of the body and the stolons. This system in the Clavelinidse was originally 
a bud-producing apparatus, and when the property of gemmation was lost (in the 
Ascidiidse) it became useful as an accessory" organ of respiration,^ and was seized upon 
and evolved by the action of natural selection into the complicated system of vessels 
and bulbs found in some Ascidiidse and Botryllidae. It is interesting to find that in the 
Botryllidse, where the property of gemmation has been acquired, the vessels in the test 
have in some cases {e.g., Sarcobotrylloides ivyvillii, see p. 59) returned to their original,' 
function of producing buds in their terminal enlargements. 
The branchial sac in the BotryUidge is well developed, and agrees with that of the 
Simple Ascidians from the point B. (fig. 11, p. 388) onwards in having well developed 
internal longitudinal bars. The reproductive organs are found in a condition which 
suggests the close relationship with the ancestral Cynthiidse, which is shown in the 
diagram (fig. 11, p. 388). 
The genus Symplegma, as I have already pointed out (p. 144), unites the characters 
of the families Distomidse and Botryllidae. In the systematic part of this Keport I placed 
Symplegma with some hesitation in the Distomidse, but I am rather inclined now to 
regard it, on account of the structure of its branchial sac and dorsal lamina, as being 
more closely related to the Botryllidse than to the Distomidse, and therefore I have 
placed it provisionally in the phylogenetic table (fig. 11, p. 388) on the termination of a 
side branch from the ancestral BotryUidse. 
For the different conditions of colony, systems, and Ascidiozooids found in the four 
genera of the BotryUidse I may refer to the systematic part of this Report (see p. 37), 
where they are described and their probable relations discussed. The ancestral 
Botrylhdse, with moderately thick colonies, probably divided into two series ; one — with 
regular stellate systems and ovate Ascidiozooids — leading to (l) Botryllus, in which the 
colony became thin and incrusting, and (2) Polycyclus, in which the colony became thick 
and massive ; and the other — with elongated irregular systems and cyhndrical Ascidio- 
zooids — giving rise (1) to Botrylloides with thin colonies, and (2) to Sarcobotrylloides 
with thickened colonies (see fig. 15, p. 398). The manner in which the systems probably 
became complicated, and in which the Ascidiozooids may have changed their form as a 
result of the modification of the systems, has already been described (see p. 40). 
1 See Herdman, On the Evolution of the Blood-Vessels in the Test of the Tunicata, Nature, vol. xxxi. p. 247. 
