MORSE: OBSERVATIONS OX LAMELLIBRANCHS. 159 
family were away and there was no vibration caused hy tlieir 
moving about and then for the first time one of the specimens 
timidly thrust out its syphons. It was extremely sensitive 
to any jar and placing a pencil ever so carefully on the table 
caused it immediately to contract. The syphons were entirely 
separate and nearly as long as the shell; the anal syphon was 
bent in a sharp curve dorsally and the branchial syphon in a 
similar manner ventrally. The syphons when full}' expanded 
are white and translucent. The opening of the anal syphon has 
six short blunt tubercles; the branchial orifice has four of the same 
shape. These tubercles terminate in faint longitudinal lines 
marking the syphons precisely as described in the English species. 
At intervals the syphons enlarge as if about to burst, becoming 
semi-globular at the ends; then, suddenly collapsing, the tubes 
become narrow and opaque white in color. The tubes may 
perform this action independently. They twist under each 
other and are very mobile. No trace of a valve was seen. The 
most marked peculiarity of T. conradi is the sharply defined 
collar which surrounds the base of the syphonal tubes. This 
collar is a prolongation of the mantle with an extension of the 
periosteum, as seen in many other lamellibranchs, but in no 
instance have I met with a description of any structure approach- 
ing the collar of T. conradi. It flares Hke the corolla of a flower 
and its edges are reflected as shown in the figure. This prolonga- 
tion of the posterior border of the mantle is seen in other forms 
but in no case with the definiteness or freedom from the base of 
the syphonal tubes as seen in this species. In Saxicava and Mya 
the prolongation of the mantle is closely adherent to the tubes. 
My own observations show that in Anatiiia papyracea the mantle 
is prolonged at the syphonal end but does not surround the tubes 
like a collar nor are its edges reflected. Pandora trilineata has a 
transparent envelope surrounding the base of the united sj'phons 
and closely adherent. In Yoldia limatida and ]'. sapotilla there 
are distinct lobes of the mantle flanking the sides of the tubes. 
In Solecurtus gihhus there are two projecting and rounded lappets 
corresponding to the separate syphons. The syphonal collar 
of T. conradi, its separateness from the tubes, and widely reflected 
edge are, so far as I know, unique among the lamellibranchs. 
Whether this feature should constitute a generic character I am 
