190 Corres2Jondence — E. U. Cunninghain-Graig. 



by a relatively narrow isthmus of sand whicti is never below high- 

 water mark in any part of its length. The size of the areas of land 

 may vary to any extent, but the highest points in both must rise 

 higher above sea-level than any part of the isthmus. The best 

 example is that known to English residents in Hong-Kong as Dumb- 

 bell Island, which consists of two granite hills, each about 300 feet 

 high and 1 J miles long, connected by a low sandy isthmus 170 yards 

 wide, rising 10 feet above normal high-water mark. The Portuguese 

 colony of Macao is, perhaps, the most typical example of a dumb- 

 bell peninsula. 



There are several varieties of dumb-bell islands, and their origin 

 is ascribed by Mr. Schofield to (1) decay of rocks, (2) partial sub- 

 mergence of the land, (3) rainstorms, (4) destruction of vegetation, 

 (5) marine action, and these causes are examined and explained in 

 detail. Nearly all the dumb-bell islands and peninsulas are com- 

 posed of granite, generally partly decayed, and their evolution can 

 be followed through all its stages in the examples near Hong-Kong. 

 The paper concludes with a list of the islands and peninsulas of 

 this type near Hong-Kong, to the number of forty-six. 



THE SGUEB OF EIGG. 



Sir, — I have no wish to enter into a controversy upon the subject 

 of the Sgurr of Eigg with Dr. Harker, whose letter, I may be pardoned 

 for suggesting, savours rather more of the " Don " than of the 

 " Survey man ". I may remark, however, that I have read 

 Dr. Harker's paper carefully, and did not find it convincing. The 

 occurrence of granite pebbles in the conglomerate may not be new 

 now ; I believe it was new when Mr. A. S. Reid and I made the 

 discovery — in 1898. I have never seen any of the older granite 

 fragments that Dr. Harker mentions, the granites " that are not 

 exposed at the surface " ; possibly they may be of Old Red Sand- 

 stone age. The granite pebbles we discovered in the conglomerate 

 are certainly not of Old Red Sandstone age. In any case, this point 

 concerning granite pebbles is arguable, but is not vital. 



The question at issue between Dr. Harker and myself is something 

 much more important, namely, the relative values of theoretical and 

 field evidence. I bow to Dr. Harker's knowledge of microscopic 

 petrology, but in matters of field evidence each man must rely 

 on his own observation. It is possible to argue at any length about 

 possibihties, probabilities, and theoretical matters, but there should 

 be no possible mistake about facts as observed in the field. I cordially 

 agree with Dr. Harker that a little knowledge is dangerous, but, on 

 the other hand, assurance as to facts is very safe. Anyone who has 

 a lengthy experience of economic geology — the kind of geological 

 work that is practically useful, as distinguished from that which 



