September 22, 1911] 



SCIENCE 



361 



to the truth. A generalization which 

 brings together a mass of scattered obser- 

 vations, and endows them with meaning, is 

 not invalidated by the discovery of excep- 

 tions. These merely prove that it is not a 

 final expression of the whole truth, and 

 may point the way to its revision and cor- 

 rection. 



Take, for instance, our provisional law 

 of the distribution of the two branches of 

 igneous rocks in defined regions. It has 

 been objected that leucitic lavas, having 

 therefore very decided alkaline or Atlantic 

 affinities, are known at several places 

 within the limits of the main Pacific re- 

 gion, where they are associated with ande- 

 sitie and other calcic rocks. Now, the only 

 area for which M'e have anything like full 

 information is the island of Java. Here, 

 according to Verbeek and Fennema, the 

 great plateau-lavas of Tertiary age are 

 exclusively of andesitic types, and the 

 same is true of the long chain of 116 vol- 

 canic centers, which represent the later re- 

 vival of activity. As against this record 

 there are five volcanoes, long extinct, 

 which at one stage erupted leucitic lavas. 

 Whether we suppose these to be aberrant 

 derivatives from an andesitic magma, or, 

 much more probably, an incursion from 

 the neighboring alkaline region, it seems 

 reasonable to regard these very exceptional 

 occurrences as of the second order of im- 

 portance, and to set them aside in a first 

 attempt to reduce the facts to order. 



The discovery of various alkaline rocks 

 on Hawaii, Samoa, Raratonga, Tahiti and 

 other islands in the midst of the Pacific 

 Ocean raises, I think, a different question. 

 So far as is known, these rocks are not 

 found in close association with characteris- 

 tic calcic types. Suess's masterly discus- 

 sion of all the geographical and hydro- 

 graphical data hitherto obtained makes it 

 clear that an Atlantic as well as a Pacific 



element of structure enters into some parts 

 of the Pacific basin. In certain areas, such 

 as the Galapagos Archipelago, the coming 

 in of the Atlantic regime is quite clearly 

 reflected in an alkaline facies of the igne- 

 ous rocks, and such exceptions are there- 

 fore of the kind which go to prove the rule. 

 Both Max Weber and Lacroix have ex- 

 pressed the opinion that the andesitic 

 branch of rocks is characteristic of the 

 border of the great Pacific basin rather 

 than the interior. It is possible that 

 further knowledge may justify this con- 

 clusion, and still only confirm the relation 

 which is claimed between the two tectonic 

 types and the two petrographical facies. 

 ileanwhile we find clear evidence else- 

 where that vertical subsidence and lateral 

 thrust have sometimes occurred in the same 

 region or in the same petrographical prov- 

 ince; nor need we go far from home to 

 learn that the complexity of structure thus 

 implied is accompanied by a corresponding 

 peculiarity of petrographical facies. 



THE NOBTH BRITISH TEETIAET PROVINCE 



In order to illustrate this point in a con- 

 crete instance, I Avill discuss very briefiy a 

 single petrographical province, viz., that 

 which occupied the northern part of Brit- 

 ain in early Tertiary times. Professor 

 Judd has regarded this as forming part of 

 a larger ' ' Brito-Icelandic province ' ' ; but, 

 while recognizing many affinities between 

 our rocks and those of higher latitudes, I 

 think that the North British area possesses 

 enough individuality to be more properly 

 treated as a distinct unit. The record of 

 igneous action here is exceptionally com- 

 plete and well displayed. Our knowledge 

 of it is derived in the first place from Pro-. 

 fessor Zirkel, Sir Archibald Geikie and 

 Professor Judd, and more recently from 

 the detailed work carried out by the Geo- 

 logical Survey of Scotland. This latter is, 



