190 BULLETIN OF THE UNITED STATES FISH COMMISSION. 



they have found appropriate places, they begin to change. One by one the bivalve 

 characters become masked, and the little bivalves are transformed into the very 

 long, worm-like shipworms which are found in wooden structures in salt water the 

 world over. 



But along with the transformation the bivalve shell is preserved, though it is 

 much modi tied as compared with other bivalve shells, and covers only a small part of 

 the head end of the body. With it the shipworm excavates the burrow in the wood 

 in which it lives, and seems equally able to penetrate the hardest or softest kind of 

 wood with equal facility. As the wood is grated away by the shell, the small particles 

 are taken into the digestive canal, and the debris is extruded through the anus; but 

 whether it serves for food in any way is a question in dispute. During its life in the 

 wood at least the larger portion of its nutrition is taken in through the tube which 

 at rest hangs in the water, and consists of small animal, and especially vegetable, 

 organisms. In thinking of shipworms, then, it should be remembered that the wood 

 in which they form their burrows is primarily for their own protection, and that with- 

 out this protection their long, naked, delicate bodies are defenseless. 



At Beaufort all kinds of unprotected wood become literally riddled in a very short 

 time. Two kinds of worms are found there in great and about equal abundance 

 Teredo norvegica and Xylotrya fimbriata, whose mode of spawning has been already 

 described. A very small proportion of specimens were of Teredo navalis, one of the 

 common European forms, in which the eggs are retained in the gills of the mother 

 during a considerable period of their development, perhaps almost till time for them 

 to >< -t into the wood. It is apparently this last species which the writer has found most 

 abundant in Long Island Sound, though a considerable portion of Xylotrya Jimbriata 

 were also found. 



The breeding season in North Carolina, so far as determined, lasts at least till 

 the middle of August and perhaps throughout the summer. That the latter is true 

 is indicated by two sets of facts. In the first place, individuals were found with ripe 

 sexual products during the early part of August, and the young derived from eggs 

 laid at this time must continue to set till September or later. In the second place, the 

 young were setting in the wood abundantly till the middle of August, a fact which 

 indicates that the same continues to some degree for some time longer. Of course, 

 from an economic standpoint, the period during which the wood is attacked is one of 

 the most vital points to discover. 



The number of young produced is amazing estimated in oqe case, from a single 

 very large female, at 100,000,000 and while the greater part are lost before the 

 setting stage is reached, yet the number that set is very great, and one of the most 

 discouraging features in dealing with shipworms in a practical way. If the spat were 

 of fairly appreciable size and set in but moderate numbers it might be feasible, by 

 the careful removal of all old piles and other old timbers, to Sufficiently reduce the 

 number to a minimum. But when, under favorable conditions, over 100 to a square 

 inch set where there is not room for more than one or two to reach maturity, it is 

 easily seen what an excess is always present and how futile it is to try to combat the 

 larvre before they enter the wood. The practical way, of course, is to prevent their 

 entrance into the wood by protecting the wood with copper paint and sheeting. 

 With small piles and timbers it would seem to be worth while to try various means of 

 keeping the bark on the wood, which, so far as I know, has not been done; for it is 

 well known that as long as the bark is on timbers they are not attacked by shipworms. 



