4 Tun. 
TUNIC AT A. 
from the yolk: of the ovum. The testa-cells, or granular cells, are also 
structures eliminated from the ovum. 
E. van Beneden & C. Julin (2 and 4) discuss the post-embryonic 
development of Phallusia scabroicles , especially the formation of the 
atrium, the stigmata of the branchial sac, the renal vesicles, the repro¬ 
ductive organs, the visceral ganglionic cord, the dorsal tubercle, and the 
circle of tentacles. The cloaca or median dorsal part of the atrium is 
formed by a depression of the dorsal surface of the larva between the 
two embryonic atrial apertures, and is therefore lined throughout by 
epiblast; while the peribranchial cavity has its inner wall formed by 
hypoblast. The primary stigmata first formed in the young Ascidian 
become broken up each into a row of secondary stigmata. The cavities 
of the renal vesicles, as well as of the reproductive organs, probably 
represent the coelom of the Vertebrata. The visceral ganglionic cord is 
derived, not from the spinal cord of the larva, but from the visceral 
ganglion. 
Ulianin (28) shows that in Distaplia the stolon arises in the larva 
before the tail is lost, and produces buds, so that a young colony is 
formed while the larva is still free swimming. The ascidiozooids are 
protandrous, the testes being produced before the ovaries. The adult 
colony, according to Ulianin, contains some ascidiozooids which have not 
been produced by gemmation, but have developed from ova through 
larvae which have remained in the common test in place of becoming 
free-swimming. 
Chabry (5) describes the segmentation of Simple Ascidians, with the 
view of supplementing the account given by Van Beneden & Julin, and 
of showing that it differs very slightly from a regular segmentation. 
Seeliger (23) continues his account of the development of Clavelina , 
and deals specially with the process of gemmation and the bearing of the 
life-history upon the phylogenetic relations of the different groups of 
Tunicata. 
E. van Beneden & C. Julin (3) have shown that in Clavelina the 
first segmentation plane corresponds to the median plane of the gastrula 
and of the future body. They have studied the details of the segmenta¬ 
tion and development of the embryo, and have corrected some errors into 
which Seeliger had fallen. The mesoblast arises in two portions—a pos¬ 
terior, which gives rise to the muscles of the tail, and an anterior, which 
forms from lateral diverticula of the archenteron (consequently the Tuni¬ 
cata must be regarded as Enteroccela) . The coelomic cavities in these 
diverticula become obliterated very early, and never undergo metameric 
segmentation. The cells of the mesoblast become scattered through the 
blastoccele, and thus form a secondary, or false, mesenchyme. After a 
detailed study of the oogenesis in Clavelina , the}^ confirm Kowalevsky’s 
original idea that the testa-cells are derived from the follicle epithelium. 
From a close comparison of the development in Clavelina and in Amphi- 
oxus, they consider that the Ascidians cannot properly be regarded as 
segmented animals, and that the body of an adult Ascidian, Salpa or 
Doliolum, is the equivalent of the cephalic segment and the first meso- 
blastic segment only of the larval Amphioxus. They conclude that the 
