CURRENTS. 243 



This fact is of some importance, for descents made over the banks on the heels of the south-west 

 monsoon showed that here and there the coarse sand was thrown into ridges, usually — almost always — 

 about 2 feet apart, 4 inches high, and facing south-west. These ridges could not be produced by a 

 bottom current, for such a current would tend to drift the sand to the north-east and toward the land, 

 the accumulated result of which over many centuries would ultimately sweep landward the whole of 

 the sand and eventually leave a clean bottom. No such result has, however, taken place. 



The comiter effect of the north-east monsoon may be entirely neglected, for not only are the 

 banks directly under the lee of the land, but they are further protected by a more or less continuous 

 fringing coral reef. 



The production of these ridges is obviously due to the surface agitation affecting the bottom layer 

 of water. But the bottom movement is vertical and not horizontal ; it is undulatory and not progressive. 

 In other words, it is a wave and not a current. 



It has been previously slio\ra that on the area under lease the fuU force of the south-west monsoon 

 is not felt owing to the penetration of the oceanic current, and consequently the bottom is almost calm. 



To the south of the plateau (on the area still under Government control) and on the west coast 

 of Ceylon south of TaUaivillu Point, where the plateau is narrow and swept by an oceanic current during 

 both monsoons, and where the water is shallow, it is most likely that a bottom current is usual. In fact, 

 our experience on these banks corroborates this idea, and a sandy bottom is rare on these southern banks. 



Referring in particular to the area under lease, the banks to the south of the plateau, and 

 including the Muttuvaratu Paars, are for the most part made up of dead coral, and sand only exists 

 close to the shore. 



In the vicinity of Karativu shoal bottom disturbance is naturally more evident than elsewhere, 

 due to the sweU breaking on the shoal. But this effect is purely local. 



On the southern extremity of the West Cheval there is also evidence of bottom disturbance. 

 Here the rock is covered in places by a few inches of coarse sand mixed with shell debris, and this sand 

 is thro^vn into ridges averaging 2 feet in length and 6 inches in height, leaving the rock bare between 

 the ridges. In all cases the coarse sand is composed of angular quartz, oxidized on its outer surface, 

 shghtly impregnated with iron, and of a yeUow appearance. The ridges all face south-west, and their 

 sides are steep, suggesting tliat they can only have been formed by vertical motion. 



Apart from the vicinity of the Karativu shoal, there is little evidence of any disturbance. The 

 rock is practically bare on the Mid-west, Mid-east, and North-west Cheval. The North-east Cheval is 

 also very free from sand. 



On other area, such as the Periya Paar Karai and Vankali Paar, where the rock is covered in places 

 by only a few inches of coarse sand, the ridges are usually 2 feet apart and up to 6 inches high. On the 

 Moderagam Paar the ridges are 3 feet apart and only 2 inches high. 



On the real sandy wastes (in contradistinction to paar ground covered only by a few inches of 

 sand) the sand, as we have seen, is fine-grained and very compact. The ridges over such areas are very 

 slight indeed, and are never more than 1 or 2 inches high and about 2 feet apart. 



Besides the evidence afforded by the ridging of the sand, which aU over the plateau is on the 

 small scale indicated, there exist numbers of crevices and holes on the floor of the ocean. In some cases 

 these holes are caused by disintegrated coral having become dislodged, whilst in other cases they are 

 natural configurations of the rock. Many of these holes are quite 3 feet deep, and one would naturally 

 expect to find that if considerable movements in the sand took place, that these holes would eventually 

 fill up. Yet they have often been examined, and never more than 4 inches of coarse sand found 

 in them. 



On the West Cheval Paar, in 9 fathoms of water, the rock has been observed to terminate 

 abruptly in places on the west seaboard with a drop of 2 feet on to a sandy bottom. Here, too, one 

 would expect to find that if the sand was carried about by a bottom current in a progressive motion, 

 the sand would silt up against such an elevation of the rock, particularly durmg the south-west monsoon. 

 Further, during the drift bottle experiments carried out over many years on the plateau, many 

 bottles sank on located positions. At least four such bottles were picked up several months afterwards 

 lying on the sand in tlie position where liberated. 



