472 BOUNDARIES OF THE SEA 



Only a fairly small percentage of the total Indo-West Pacific 

 coastal fauna has reached Hawaii in spite of the chain of "stra- 

 tegically" placed islands along the southern border of the Kuroshio 

 current, islands which might have been reached by several larvae 

 at normal current velocities. This tends to show that in order to 

 be of true importance as an intermediate station for larval trans- 

 port, the areal extent of an island or island group must not be too 

 small. It must be remembered that most lar\'ae aggregate in the 

 plankton; they are transported in swarms. If an island is very 

 small, huge larval swarms may pass it a short distance away 

 without contacting it. On the other hand, this aggregation may 

 also prove extremely useful for the lar\'ae. In several marine 

 invertebrates a whole population will spawn epidemically (Thor- 

 son, 1946, 1950). Such a swarm of larvae, all from the same brood, 

 is caught up in and drifts with the current, exposed to much the 

 same conditions in the same water mass. In most cases the whole 

 swarm will perish before reaching a new transoceanic coast. But 

 if, for instance, once in some two hundred years, they are successful 

 in crossing an ocean, then a whole swarm may arrive — enough to 

 found a new population. 



Summary 



Eighty per cent of all bottom invertebrates with pelagic larvae 

 have a planktonic life of less than 6 weeks. Even if the length of 

 this stage is doubled or if the larvae are transported by especially 

 rapid currents, they never seem to ha\'e even the slightest chance 

 to cross the larger ocean basins. 



The group that has the least chance for long-distance transport 

 is the lamellibranchs. Polychaetes and echinoids include at least 

 some species which have a fairly long stay in the plankton, 

 although they cannot be regarded as true long-distance larvae. 

 Some tropical ophiurans and perhaps also some tropical asteroids 

 may have chances to cross even the widest ocean basins, provided 

 that Mortensen's observation on the "budding larval polyps" 

 holds true. True long-distance larvae (i.e., larvae directly "built" 

 for long-distance transport) seem only to occur in special groups 

 of prosobranchs and decapod crustaceans. They are equipped to 



