DISPEESAL OF LIFE VISHER 317 



Queensland of certain forms known in New Caledonia. Both the 

 normal trades and numerous hurricanes pass over New Caledonia 

 and thence to Queensland, and yet the coconut is established at only a 

 few points on the Queensland coast, and certain species of land snails 

 abundant in New Caledonia are unknown in Queensland. 



The general lack of the coconut in Queensland appears not to be 

 due, however, to a lack of seed there, for they are often noticed 

 sprouting along the shore. The severe droughts during the cooler 

 season of each year, characteristic of this part of Australia, may 

 be the factor preventing their widespread establishment on this 

 coast. 



The absence from Queensland of the large showy land snail {Pla- 

 costylus), common on the islands to the eastward, may be due to 

 the presence in Australia of enemies that prevent their establish- 

 ment. The Australian bush turkey feeds upon similar Mollusca, 

 and it has been suggested by Curator Charles Hedley, of the Aus- 

 tralian Museum, as possibly responsible for the absence of this island 

 snail in Queensland. 



HURRICANE FLOODS AND LIFE DISPERSAL 



Another way in which hurricanes have influenced the dispersal of 

 land forms is in connection with the floods they cause, whenever 

 excessive amounts of rain fall in a short time. There are numerous 

 records of the fall of more than 10 inches in two days, and some rec- 

 ords of more than 60 inches in three days.^^ Under such conditions 

 normally rather small streams become rivers and carry to sea vast 

 quantities of driftwood. The river banks are eroded badW, and 

 many trees are undercut and carried out to sea. During the excessive 

 rains, large masses of dirt and loose rock upon steep hillsides often 

 slip or slump, sometimes temporarily damming valleys. In case the 

 dam breaks, the sudden rush of waters does its part to contribute 

 natural rafts of driftwood, with their load of land animals and 

 seeds. Furthermore, the animals are not nearly so likely to be picked 

 off a raft by sea gulls or other birds during a hurricane as they are 

 in ordinary weather. Hence, the absence of long rivers flowing into 

 the Pacific, with the exception of the Chinese rivers, should not lead 

 to the assumption that natural rafts of considerable size and bio- 

 logical dispersing possibilities are lacking in the Pacific." 



1^ Visber, S .S., " Tropical climates from an ecological viewpoint," Ecology. Vol. IV, 

 pp. 1-10, 192.3, and " Tropical cyclones of the Pacific," Bull. 20, Bishop Museum, Hono- 

 lulu, 1925. 



" Pilsbry, H. A., (loc. cit.) has made this erroneous assumption. He states : " The 

 argument for distribution of animals by natural rafts has never been more convincingly 

 stated than by W. D. Matthews In his paper ' Climate and evolution.' Much of his argu- 

 ment, however, is not applicable to the Pacific islands. Here we have no large rivers to 

 give forth natural rafts. If a single tree Is washed to sea It must be very exceptional." 



