302 THE COMPOSITION OF ROCKALLITE. [June I914. 



DlSCUSSIOX. 



Prof. W. W. Watts commented on the exhaustive character of 

 Prof. Judd's account of rockallite from the very scanty amount of 

 material at his disposal. The speaker referred to the strange fact 

 that the rocks of so many of the islands off Britain showed 

 characters unrelated to those of any known rocks on the mainland. 

 He cited especially, in this connexion, Ailsa Craig, Lundy, the 

 Wolf Rock, Bear Island, and Rockall. 



Dr. J. W. Eyais'S referred to the fact, to which Dr. Harker had 

 drawn attention in his Presidential Address to the Geological 

 Section of the British Association at Portsmouth in 1911, that 

 alkaline rocks were apt to occur on the margin of igneous areas. 

 Those of Kathiawar, for instance, were found in an outlying portion 

 of the basalts of the Deccan Trap : thej might be compared with 

 the Rockall intrusive, which was associated with basaltic rocks, 

 found by dredging in the neighbourhood, that might (as Prof. 

 Grenville Cole had suggested) be of the same age as those of the 

 North-East of Ireland and the Inner Hebrides. The earlier 

 accounts of Rockall which attributed to it a larger area, and in one 

 case mentioned another island in the immediate neighbourhood, 

 might be explained on the supposition that a portion of the basaltic 

 platform had once emerged above the surface of the sea, and had 

 since been removed by marine erosion. 



Mr. F. P. Mexxell said that he was interested in the Author's 

 suggestion concerning the presence of cerium in the acmite. His 

 own experience in Tropical Africa showed that cerium commonly 

 occurred in the granites as the silicate orthite, or allanite ; but 

 there was another occurrence, interesting in the present connexion. 

 Near Bulawavo a rock consisting essentially of microcline and 

 augite occurred. The latter mineral was not aegirite or acmite, 

 but was evidently an alkaline variety — as it was pleochroic, and 

 had an extinction-angle smaller than that of most pyroxenes. In 

 some altered varieties of the rock it was changed to epidote, and 

 certain offshoots were practically transformed into felspar-epidote 

 rocks. The epidote contained brown patches of orthite, which 

 must have been formed from substances originally present in the 

 ferromagnesian mineral. 



He noticed that the Author did not regard geological occurrence 

 as an important factor in rock-classification. But, after all, this 

 was the real difference between, for instance, granite and rhyolite. 

 Those who worked at igneous rocks in the field would, he thought, 

 be sorry to give up such intelligible principles as occurrence in 

 favour of purely artificial systems like the Quantitative Classifi- 

 cation of wdiich Dr. Washington was joint author. 



Mr. R. S. Hermes thought that the fact that large islands had 

 been placed, in accordance with reports of early navigators, on the 

 Rockall and other banks was probably due to clouds settling upon 

 the banks and causing at a distance the appearance of land. He 

 reminded the Fellows of a landing on the island, effected some few 

 years ago, when a liner from North America ran on to it and was 

 wrecked with loss of life. Had there been any geologists among 

 the survivors, more specimens of the rock might then have been 

 obtained. 



