320 



By whatever means its distribution may be accomplished, the infection of 

 untreated wood exposed in places where Limnoria is known to occur is very regular 

 and rapid. Test timbers exposed at several localities in the lower bay have uniformly 

 been attacked within one month, which is the minimum length of exposure of our 

 test pieces. This has been the case during every month of the year. These test boards 

 when exposed further from two to twelve months showed regularly increasing sever- 

 ity of attack following the initial infection. 



Excluding the somewhat problematical distriluiti(jn In- shipping, it was assumed 

 that Limnoria may be distributed either as separate free-swimming individuals, or 

 collectively, attached to driftwood and inhabiting burrows in its surface. 



Among the factors which would be important in distribution of separate indi- 

 viduals, special attention was given to their swimming activities, their reaction 

 towards wood and wood extracti\'es, and their abilit\' to exist without a supply of 

 wood for a consideralile time, assuming it to be the main source of their food. 



If Lirnnoria acts as its own agent of dispersal one would expect to find some 

 method of locating and recognizing wood under water rather than dependence on 

 chance contact therewith. In the case of teredo larvae, Harington (1921) has obtained 

 e\'idence that there is a wood extractive to which these organisms are positively 

 chemotropic. He obtained this extracti^•e not only with ether and alcohol, but also 

 with sea water; indicating therefore that such an extractive may be an effective 

 agent in inducing the settlement of these larvae on wood. The case of Limnoria is, 

 however, somewhat different from that of Teredo. It is necessary for teredo larvae 

 to locate wood on which they may settle within the limited period of their larval life. 

 The condition in Limnoria is nearly the reverse. The very young, recently hatched, 

 organism does not swim at all, but gradually acquires this ability as it grows. At any 

 time in the entire adult life, it may be transported to new locations and, so far as 

 we know, there is no critical or limited period in which the organism must find its 

 new location in order to survive. So, while a chemotropic mechanism for locating 

 wood would be an aid in the dispersal of Limnoria, it would not appear to be absolutely 

 essential. 



Animals li\-ing under obser\ation in acjuaria have not shown any definite chemo- 

 tropic reaction toward wood. Specimens introduced into an aquarium containing 

 l)locks of wood swim about in a random fashion and do not exhibit any behavior which 

 it would be possible to interpret as a tropism or sensitivity to the presence of the nearby 

 wood. They may come within a fraction of an inch of the wood and continue in their 

 course without settling upon it. It has been observed, however, that water-soaked 

 wood is more likely to be attacked than is wood recently introduced to the aquarium. 

 It is difiicult to tell whether this fact has a chemical or a physical basis. A similar 

 reluctance to attack fresh wood has been observed in the case of teredo larvae. 



In an experiment to determine how long Limnoria can exist without wood, some 

 organisms were kept ali\'e in a dish of sea water for over six weeks. At the end of 

 thirty days some of these animals were introduced into another aquarium containing 

 some pieces of well soaked wood. It was surprising to find that they exhibited no 

 special a\'idit\- for the wood and no swimming reactions which showed a sensitivity 

 to its presence. Some were then drawn into a pipette and placed on the wood. After 

 exploring its surface they swam away from it. This test was repeated several times 

 with the same result. The animals did not attack the test piece any sooner than 

 those organisms which had recently been removed from their burrows. It would seem 

 that if Limnoria utilized a chemical sense to detect wood it would be particularly 



