64 Proceedings of Royal Society of Edinburgh. [sess. 
which Phoronis differs from the others, namely, the absence of a 
notochord and gill-slits. 
We may first take into consideration the notochord. 
This, though an important organ in Balanoglossus, is greatly 
reduced in Cephalodiscus and in Rhabdopleura. In all three it is 
intimately connected with the proboscis (syn. epistome or pre- 
oral lobe), and shares its degree of prominence with this organ. 
In the case of Phoronis , we already know from its ontogeny 
that its larval form, Actinotrocha , has a very well-developed pre- 
oral lobe, but that this atrophies almost entirely, leaving only a 
vestigial epistome (Caldwell). This being so, we must seek for 
the notochord, not in the adult Phoronis, but in the Actinotrocha 
just before its metamorphosis. 
To prophesy is a dangerous pursuit, especially in morphology, 
but I cannot help suspecting that subsequent investigations will 
reveal some trace of an organ homologous to the notochord in the 
Actinotrocha stage. 
Such a discovery, although convincing, is not, however, essential 
to our contention, for a notochord may never have been evolved 
in Phoronis, and yet its alliance with the Hemichordata be 
unassailable. 
Secondly, there are no gill-slits. This is a feature which, so 
far as is known, is not found at all in one of the Hemichordata , 
namely, Rhabdopleura , and the gill-slits differ in number in 
structure, and even, perhaps, in function in the other two. Apart 
from this, we have pointed out that the “ branchial fissure ” in 
Phoronis performs the function of the gill-slits of Cephalodiscus ; 
and although, of course, not morphologically equivalent, yet its 
presence points out a factor which might easily account for the 
non evolution or atrophy of true gill-slits in the gut region. 
We may assume that the reduction of the epistome, consequent 
upon a sedentary life, enabled the outer row of tentacles to be car- 
ried across the median line ventrally, and thus formed the branchial 
space above referred to, which physiologically can be compared in 
every detail with the pharynx of Tunicata : the same comparison 
holds true for the branchial fissure and the atrial space. Phoronis 
having therefore, through loss of function by the pre-oral lobe, been 
enabled to evolve a complete branchial and food-collecting area (by 
