288 



recuperation before breeding activities are begun. The critical conditions affecting 

 the opening of the breeding season, accordingly, are not those found during the summer, 

 but are a function of the length and severity of the period of stress of the preceding 

 winter. 



Regardless of the time at which the breeding season l^egins, it terminates generally 

 throughout the bay in December. Occasionally a straggling settlement of larvae may 

 continue into the early part of January. Piles driven at Crockett in January, 1920, 

 were found to be attacked by Teredo and penetrated to a depth of two inches by 

 May. This indicated conclusively the settlement of larvae within that period. Also 

 on January 7, 1924, a few minute larval punctures were noted on a test timber at 

 Crockett which had not shown any evidence of attack when examined on December 24. 

 These two are the only cases that have indicated an>- lar\al settlement whatever 

 after December. 



Post-Larval Development 



The larval Teredo, having found attachment on a suitable piece of timber, begins 

 at once to undergo important changes. The shell becomes rapidly altered and adapted 

 to the function of boring in wood, while at the same time the internal organization 

 of the animal becomes modified in connection with the new mode of life and the change 

 from a diet of minute plankton organisms to a diet consisting for the most part of 

 wood. 



It has been objected to the theory we have attempted to establish (Chapter 15) 

 of mechanical boring by means of the shell, that the newly attached larva could not 

 conceivably be able to bury itself in wood by the action of the shell alone, which in 

 the larva is more fragile than paper. As a matter of fact, however, no actual boring 

 is done until after the shell has become in a measure adapted to such work. 



The surface of submerged timber is invariably coated with a film of surface 

 debris, composed of silt, diatoms, algae and miscellaneous detritus. The attached 

 teredo larva has at first only to scrape about in this surface debris until it has worked 

 itself in and come in contact with the surface of the wood. This much is readily 

 accomplished by the larval shell and foot, but further progress must await the de- 

 velopment of toothed ridges on the shell. We have never seen any larvae even partly 

 imbedded in the wood which had not developed one or more series of such ridges. 

 So rapid is the addition of ridges at this critical period in the life of the organism that 

 before it has buried itself from view at least a half-dozen ridges have been added, 

 the shell in the meantime having very nearly assumed the form it presents in the adult. 



The method by which the shell becomes modified from the simple protective 

 covering of the pelagic larva to the specialized boring tool of the adult is extremely 

 interesting. Growth of new ridges does not occur evenly around the margins of the 

 larval shell, but occurs most rapidly at the antero-ventral margin. Thus the new 

 ridges curve outward ventrally, in the region of more rapid growth, and this condition 

 becomes more and more accentuated with the addition of successive ridges, until the 

 later ridges are bent at almost a right angle. This angle marks the division between 

 the anterior lobe and the median lobe of the adult shell (see p. 196). The two segments 

 of each ridge, as marked ofT by this angle, are morphologically one. The formation 

 of this angle accounts for the anterior gap through which the foot is protruded. 



In the meantime there has been a rapid deposition of shell at the postero-ventral 

 margin of each valve, forming the auricle, which increases rapidly in size to provide a 

 sufficient surface for the attachment of the rapidly growing posterior adductor muscle. 



All this is accomplished, and even rudimentary pallets are developed, before the 



