December 20, 191 7] 



NATURE 



315 



little to the east of the Dutch-German boundary. It 

 is regarded by the author as a place of exit for the 

 northern migration from Indonesia, and he quotes in 

 full an account of the people by Neuhauss. A 

 vocabulary in the latter's work, " Neu-Guinea " 

 (probably derived from the trader Schulz, described 

 by Neuhauss as a "drunkard" and by Friederici as 

 "a man of no particular intelligence"), is, when cor- 

 rected by a shorter list of Friederici 's, the basis of 

 Mr. Churchill's argument. 



Although the words discussed are so few (about 34), 

 the author belittles the existing Melanesian vocabula- 

 ries, and suggests (without authority) that others — 

 Codrington, Ray, and Friederici — have based their 

 conclusions on these scanty lists, 



Mr. Churchill totally ignores the structure of the 

 languages, and does not distinguish between 

 Melanesian and non-Melanesian languages. His lists 

 of cognate words are designed to show their varia- 

 tion from simple stems which the author finds in the 

 Polynesian forms. Thus makan, "eat," is derived 

 trom ani; tebii, "sugar-cane," from to. Such deriva- 

 tions are impossible in Indonesian philology. The 

 former presence of Polynesians throughout Indonesia 

 is regarded as established by the tables. The author 

 supposes them to have been expelled from the archi- 

 l>elago by a more cultured people from the mainland 

 of Asia, who assumed some elements of the Polynesian 

 speech. , 



Mr. Churchill's comparisons of Melanesian and 

 Polynesian words are open to the same objection as 

 those from Indonesia, for he will not admit that 

 Polynesian is secondary to Melanesian. He traverses 

 Friederici 's suggestion of a Melanesian migration 

 round the eastern end of New Guinea through Vitiaz 

 Strait {cf. Nature, December 5, 19 13). The whole 

 of the eastern part of the south coast of New 

 Guinea is regarded by the author as belonging to 

 Torres Straits, and the languages there are said to 

 bf more Polynesian towards the west.' For the 

 western part (t.^. the real Torres Straits region) he 

 says there are no records, and hence he supposes that 

 Polynesian survivals may be found there. This is to 

 support the contention of a Polynesian migration 

 through Torres Straits. But all the languages of 

 New Guinea from the narrows of the Straits to the 

 first Melanesian settlements at Cape Possession are 

 known, and show no Polynesian survivals. 



As to the island region of Melanesia, Mr. Churchill 

 quotes from Dr. Rivers's "Melanesian Society" th« 

 chapter on kava and betel, but will not allow a 

 possible origin of the kava culture in Melanesia. 



Mr. Churchill's book is interesting, especially in its 

 descriptive parts, and his views as to the directions 

 of the migrations are clearly set forth in a series of 

 charts. He does not disguise the difficulties of his 

 thesis, but his desire to establish the pre-eminence of 

 Polynesian among the languages of the Pacific has 

 caused him to overlook many details which seriously 

 militate against the soundness of his argument. 



Sidney H. Ray. 



CANADIAN EXPLORATION. 



SOME information on the work of the Canadian 

 .Arctic Expedition additional to that alreadv pub- 

 lished appears in the Geographical Review for October 

 (vol. iv., No. 4). The whole of the coast-line from 

 Cape Bathurst to the eastern end of Coronation Gulf, 

 except Cape Parry peninsula, was surveyed on a scale 

 of ten miles to an inch. The surveys were extended 

 to include Croker River and Rae River. Other sur- 

 veys included much of the Mackenzie delta. The most 

 important geological research of the southern party, 

 with whose work this paper bv Mr. R. M. .Anderson 

 NO. 2512, VOL. 100] 



deals, was the investigation of the copper-bearing 

 I rocks in the region of Bathurst Inlet. They occur on 

 I Banks Peninsula and most of the islands to the north 

 I of it. Though the ore so far discovered is of low 

 j grade, Dr. O'Neill, tjhe geologist of the expedition, 

 estimated the amount of ore at more than two billion 

 I tons, and believes that it can be profitably utilised. 

 ! Water transport round Alaska would be long and un- 

 certain, but a short railway to Great Bear Lake would 

 probably solve the difficulty of export. The article is 

 i illustrated with two sketch maps and a number of 

 ; valuable photographs. 



The Geological Survey of Canada has published, as 



' Memoir 84, a volume by Mr. C. Camsell on an ex- 



: ploration of the Tazin and Taltson Rivers in the North- 



I West Territories. The exploration of this region be- 



! tween Lake Athabaska and the Great Slave Lake was 



i undertaken by the author in 19 14 by a canoe traverse 



from south to north. A compass survey was made 



and checked almost daily by sextant observations. The 



report is mainly geological and is accompanied by a 



geological map on a scale of 8 in. to a mile, and by a 



I number of good illustrations. Neither the Tazin nor 



the Taltson proves to be navigable, except for short 



stretches, on account of falls and strong rapids. The 



country abounds in lakes. The lack of soil, apart 



from climatic difficulties, makes agriculture impos- 



' sible. The inhabitants are a few nomadic Indians 



I of the Chipewyan and Dogrib stocks. 



! EFFECTS OF STORAGE UPON COAL. 



A FURTHER contribution to the study of the effect 

 I ■^*- of storage on the properties of bituminous coals 

 I has been issued from the Engineering Experimental 

 Station, University of Illinois (Bull. 97). Prof. S. W. 

 Parr, who has been associated with Wheeler, Barker, 

 and Kressman in a series of experiments started in 

 19 10, summarises the conclusions arrived at, and re- 

 cords further investigations ; more particularly of in- 

 terest are those on the action of pyrites and the tests 

 of weathered coals under boilers. The general conclu- 

 sions that freshly mined coal has a large absorptive 

 capacity for oxygen, the degree being dependent on 

 the character of the coal ; that the rate of absorption 

 depends upon fineness of division and temperature ; that 

 such oxidation leads to slight increase of temperature, 

 and if not radiated more rapidly than generated the 

 action accelerates to a dangerous point (180° F. is 

 named), have for some time been recognised as the 

 main causes of spontaneous ignition. Parr concludes 

 th^t the well-recognised loss of thermal value on 

 storage is more ajjparent than real, being largely due 

 to increase of weight due to oxygen absorption. 



Much difference of opinion exists as to the part 

 played by pyrites (FeS,) in promoting heating. Experi- 

 ments are recorded showing the increase of soluble 

 sulphates in coal of various grades of fineness on 

 storage. Fine grades! show marked increase, but 

 only in one case of coal passing a ten-mesh sieve was 

 any increase noted. It is concluded that to lead to 

 sulphur oxidation two conditions are necessary : fine- 

 ness of division and presence of moisture. It had 

 been previously shown that the oxidation of 05 per 

 cent, of sulphur produced sufficient heat to raise the 

 coal, not allowing for radiation losses, about 125° F. 

 Finely divided pyrites, thetefore, may well materially 

 assist in promoting heating by itself, raising the tem- 

 perature to such a point that oxygen absorption is 

 greatly accelerated. It is shown that whilst the heat- 

 ing value of stored coal is materially reduced, such 

 weathered coal gives over-all boiler efficiencies as high 

 as fresh coal, provided a higher draught is used and 

 the fire kept thin and clear of the water-back of the 

 grate, otherwise clinker trouble is experienced. 



