TEREDO. 149 



commonly to bore in the direction of the grain of the 

 wood^ but occasionally it bores across the grain, to avoid 

 the track of any of the others/' Although this is the 

 direction which it usually takes, it is by no means un- 

 common to see perforations inclined at various angles, 

 and sometimes even made right through a tough knot 

 in a piece of oak. Montagu also remarked, with his 

 usual acuteness, that '^ the Teredo bores across the 

 grain of the wood as seldom as possible ; for after it 

 has penetrated a little way, it turns and continues with 

 the grain, tolerably straight, until it meets with another 

 shell, or perhaps a knot which produces a flexure ; its 

 course then depends on the nature of the obstruction ; 

 if considerable, it makes a short turn back in form of a 

 syphon, rather than continue any distance across the 

 grain/' The same kind of siphonal bend takes place 

 when the piece of wood, being shorter than the average 

 length of the Teredo, is nevertheless broad enough to 

 admit of its abruptly turning and doubling like a coursed 

 hare. If the space is not sufficient for its complete 

 development, the Teredo shuts itself up and closes the 

 front with a cap-shaped epiphragm; it never pene- 

 trates that end of the wood, so as to make the canal 

 pervious. The Teredo possesses the same cartilaginous 

 styliform process which I noticed in the account of 

 P ho las. The imbricated plates, or septa, that line the 

 neck of the sheath in probably every species, serve as 

 ledges to support or strengthen the pallets, which are 

 withdrawn further into the sheath as the Teredo increases 

 in length and bulk ; the last formed plate is consequently 

 innermost. Fischer counted twenty-five of these plates 

 in a sheath of T Norvegica. I do not agree with him 

 in believing that the Teredo goes on perforating the 

 wood beyond what is required for its habitation, nor 



