1923] Dore-Miller: The Digestion of Wood by Teredo navalis 387 



Analysis of wood particles removed from the excised caecum would 

 be open to criticism on more than one count. Aside from con- 

 sideration of the almost impossible number of teredos that would be 

 required for the accumulation of the necessary amount of the ingested 

 wood, we are ignorant of the length of time the borings remain in 

 the caecum before being ejected to the outside, and a great deal of 

 the material thus secured might represent only the earliest stage of 

 the digestive process; further, there is the possibility that some 

 digestion takes place after the wood leaves the caecum, during its 

 passage through the rather long intestine, especially the enlarged 

 proximal portion, which, contains a typhlosole. It seems extremely 

 desirable that the material analyzed should be secured after it has 

 passed entirely through the digestive tract in normal fashion, and 

 been ejected through the excurrent siphon. 



EXPERIMENTAL 



At the marine laboratory maintained by the San Francisco Bay 

 Marine Piling Committee on the Oakland mole of the Southern Pacific 

 Railroad, through whose cooperation these studies have been made 

 possible, for more than a year the writers have had under observation 

 in aquaria a considerable number of Teredo navalis, which highly 

 destructive species occurs locally in great numbers (Kofoid, 1921 ; 

 Kofoid and Miller, 1922). The animals have continued boring and 

 appear to thrive under conditions of aquarium life. Prom time to 

 time wood borings were ejected in considerable quantities from the 

 siphons (pi. 18, fig. 2). Such borings were carefully collected with 

 a long pipette and dried. In about eight months sufficient material 

 was accumulated by this method to permit of successful analysis. 



Two aquaria, which we will term No. 1 and No. 2, supplied all the 

 wood particles used. The original wood in both cases was Douglas 

 fir, consisting in No. 1 of squared blocks planted in the bay for 

 experimental infection and later placed in the aquarium, and in No. 2 

 of a section of piling infected at the time it was pulled. The material 

 from the two aquaria has been kept separate and subjected to parallel 

 analyses, as below described, thus rendering the results doubly 

 significant. 



The possibility was recognized of changes occurring in the com- 

 position of the wood subsequent to ejection, especially as it was not 

 always possible to collect the borings with desirable promptness. In 

 order to test the probability of such changes, some of the borings 



