161 iJEOoKns ni' thf austkalian mtskum. 



kinds, T". (lerii:i(i, T. ijiya^, T. miitira and T. sijiiKmosit togethei" witli 

 Hippopnti nifiv lie Icx^se and quite unattached to the surface eithei' by foot 

 or byssus. In my experience I'ridncna are never found on sand or mud, 

 but only on ooi-al. 



Tlie first glimpse that a traveller has of a living, boring Tr'nhicnd is 

 usually' an appai'ition of a brilliantly eolouied, round seipentine worm, 

 six or eight inches long, lying in dead coral. If he attempts its capture, 

 the worm vanishes with a click, and the locked teeth of the valve margin 

 giin at him in its place. 



A lump of Astrean coial is hei*e sliown (PI. xxxii., tig. lO) in which 

 a specimen of T. crorpK, about* three inches long hns sunk so deeply tliat 

 the upturned edges of the valves aie level with the surface of the rock. 

 It was collected at Dunk Island, Queensland by Mr. E. J. Banfield who 

 figured it in "My Tropic Isle," 1911, p. 126. The summit of the block 

 appears to have been dead when the 'rriihiruK settled thei-e. Since then 

 the coral colony grew up from the base again, the polyps intruded on the 

 anteiior end of the cell and were nbraded by the foot of the mollusc. 

 Boring Trlihiriin are always, as far as my memory serves me, found in 

 dead, not in living, coral and that a sedentary Triilnciia might become 

 imprisoned by being enclosed in living coral would, I think, be a rare 

 accident. The burrow is smooth inside and is lai'ge enough to allow 

 considerable movement to and fro and of opening the valves fiom side to 

 side. But the enti-ance is narrower than the chamber both in length and 

 breadth, so that even a. single valve cannot be withdi'awn through tlie 

 opening without breaking it. 



All early stage in burrowing is repiesented by Plate xxxiv., fig. l.'-l. 

 Hei'C a voung 7'. i-rarea, .35 mm. long has begun to operate on a pebble of 

 dead and much woi'n coral. I gathered this specimen on the beacli at 

 fri-een Island, (^)iu'ensland. It well illustiates the fact that the cell of a 

 buried 'rn'thii-ini is fotmed not by the envelojnneiit of gi'owing coral but 

 bv the [tenet I al ion of (he mollusc into dead coral. In this case the burrow 

 is driven obliquely, the left valve being dee[)ei' sunk than the right and 

 the doi-sal deeper than the ventral end, about one third of the left and a 

 sixth of the right valve are subTueiged. The dorsal posterior angle is 

 lifted clear and the anterior ventral angle is just covered. At this stage 

 the bivalve could probably \vithdra\v and commence another hole, but 

 fuilher excavation would entail imprisonment for life. Under denudation 

 the dead cural melts jiwav lapidlv. so that whether the mollusc descended 

 into the rock pcrpeiidicu larl v oi' obliquely the entiance to the cell would 

 l)e removed liv wcat liciini;' of the rock sui'face and the route lost. 



A young specimen of 7'. i-rnri'ii, \S mm. in length (PI. xxxiv., tig. ll), 

 has a pedal opening of .'> x S mm. Hut in a s[iecimeii ."11 mm. long this 

 aperture has inci'cased dispi oportionatel v, being IT) x 7 mm. (PI. xxxiv., 

 tig. I")). l-*rol)!ibly this imlicates that the lai'ger shell hail reached a time 

 ot lite in wliicli liuiiowing became of gi-eater im|)ortance. 



