436 TRANSACTIONS OF THE WAGNER FREE 



To the students of the fossils a few words of caution may be useful, 

 especially as several excellent paleontologists have at times been led into error 

 by unfamiliarity with the facts to be gained by a study of the recent forms. 



Stoliczka has divided the group into a number of genera and two sub- 

 families — a proceeding which future researches may justify; but, from an exami- 

 nation of a large number of recent species, I am led to suspect that the 

 differences upon which they are chiefly based, such as the supposed forms of 

 the foot in Siplioiiodentaliiim and Dentalium proper, are less important than 

 one would suppose from the few published figures, and require further inves- 

 tigation before they can be safely used in systematic work. At any rate, for 

 my present purpose, I prefer to retain the old nomenclature. The researches 

 of Sars have shown that the arched side of the shell is to be regarded as 

 ventral and the concave side dorsal, but to avoid confusion in comparisons I 

 have not corrected here the terms in general use, which are the exact opposite 

 of these. 



The posterior end of the shell in Dentalium may be circular or ovate and 

 evenly truncate, oblique, undulated, fissured in the median line, or with slits, 

 lateral or ventral, or both. 



The most usual form is that which is evenly transversely truncate ; the 

 next most common style is one in which there is a dorsal wave, or sulcus, or 

 even a narrow slit of some length. The other combinations are less common. 

 In a wide sense these variations depend upon the form of the mantle-edge or 

 internal lining of the shell, which is an extremely contractile membrane 

 capable of secreting shelly matter. This, like the shell, may be simply 

 tubular, sulcate, undulated, etc., and when the shell is absolutely perfect the 

 posterior end reflects the form of the membrane which secreted it, and which 

 is known from observations on the recent shells to be capable of repairing 

 damages to the calcareous tube which protects it. 



The modifications of this part, therefore, are, as a rule, merely of specific 

 value, and even in some cases are known to fluctuate widely in different indi- 

 viduals of the same species. The genus Sipliouodcntaliiim is said to have its 

 posterior end always more or less fissured, but few of the supposed Siphono- 

 dentalia have had their soft parts accurately recorded, and the variability in 

 Dentaliuni leads one to doubt if the rule is necessarily invariable in the other 

 group. In Cadulus the orifice is known to differ between different species. 



Another modification of the orifice has given rise to much misconception. 

 Species with very thin shells usually live buried in soft mud which measur- 

 ably protects them, but others with heavy shells appear to be more versatile ; 

 at all events, if the small end of the shell is accidentally broken off, the animal 

 can repair it, and in species which have a simply tubular mantle and a thick 

 shell the repairs take the shape of a small tube projecting from the blunt end 

 of the large one, as it is impossible for the mantle to secrete a shell which is 

 as large and thick as the original at the point of truncation. I have examined 



