195 



other groups of moUusks affords, if not an adequate basis, at least a useful basis for 

 specific distinctions, is in the case of the Teredinidae only occasionally useful for such 

 purposes. This structure, l)eing specialized for boring in wood, is in important features 

 essentially the same in a number of different species, and in minor features the intra- 

 specific \'ariations often cover as wide a range as the inter-specific differences. 



The earlier students of these mollusks were more prone to recognize these difficul- 

 ties than .some recent workers. Tryon (1862) stated: "Another difficulty in the study 

 of the Teredinidae is the great variation of the individuals in size, proportions, and 

 markings, making an accurate diagnosis a simple impossibility and compelling us to 

 rely on a general accordance with descriptions in the most material points." He 

 further quotes Hanley. "There is one fact with regard to the shipworms, which has 

 rendered their investigation peculiarly laborious, namely, that no reliance can be 

 placed upon the relative proportions of their several parts for specific definition . 

 Hence it is absolutely necessary to examine very numerous examples in order to elicit 

 the real and permanent specific characters, and the valves alone are rarely adequate 

 for the determination of the species." 



Of possibly greater value for systematic purposes than the shells of Teredinidae 

 are the pallets. These are paired calcareous structures, peculiar to this family of 

 mollusks, which are developed at the posterior or siphonal end for the purpose of 

 closing the orifice of the burrow when occasion arises for so doing. Considerable 

 individual variation also occurs in these structures, and caution must accordingly be 

 exercised in their use as a basis of specific distinctions. 



The range of variation occurring in the shells and pallets of Teredo navalis in 

 San Francisco Bay and some of the causes which produce such variations have been 

 discussed by Miller (1922 and 1923). This will be referred to in more detail in the 

 later discussion of that species (see p. 432 ff.). The nature and range of such variations 

 in this and other species may be observed from a study of figures 106-109. 



It is obvious that if minor variations in the shell and pallets were regarded as 

 having systematic importance, there would be no end to the describing of new species 

 and the already badly tangled taxonomy of the group would be thrown into greater 

 and greater confusion. It is to be hoped that ultimately sufficient data may become 

 available to permit a revision of the classification of the Teredinidae based primarily 

 on anatomical considerations. In the meantime our only refuge from taxonomic 

 chaos is a conservatism in the evaluation of differences in shells and pallets, and a 

 frank disinclination to recognize new species except on the most decided evidence. 



Terminology I'sed i\ Description 



A brief explanation should be introduced at tliis point of the terms used in de- 

 scription of shells and pallets. 



The shell is a reduced, sub-globular structure, gaping widely in front for the 

 protrusion of the foot, and behind for the backward extension of the body. It is made 

 up of two valves which articulate dorsally and ventrally by specialized knobs. Each 

 valve is made up of three rather distinct lobes, an anterior, a median, and a posterior. 

 The last named portion is frequently referred to as the auricle, a term which will be 

 used throughout this discussion in order to simplify so far as possible a confusing 

 terminology arising from the fact that the median lobe of the shell must be further 

 subdivided into three areas, an anterior median, a middle median, and a posterior 

 median. The areas designated by these terms are indicated in figure 68. 



The ridges of the anterior lobe are fineh- serrate, while those of the anterior median 

 are coarsely denticulate; the ridges of these two areas are nearly at right angles to 



