SCIENCE. 



[Vol. XIX. No. 465 \ 



these papers, as determined by delicate chemical tests, is due 

 entirely to accidental impurities. Some of the papers that were 

 thought to have caused poisoning had been on the walls for thirty 

 or forty years. Supposing, for the sake of argument, that there 

 were sixty square yards of paper in a room, each yard containing 

 one grain of arsenic — the amount found in several of the cases 

 quoted— and that during a period of thirty years all the arsenic 

 had left the wallpaper and had entered the human system with- 

 out any being lost, this would be at the rate of one grain in six 

 months, or only 1^5^ of a grain in each twenty-four hours. Many 

 distinguished scientists have independently investigated this sub- 

 ject of poisoning from arsenical wall-paper, and they all agree in 

 saying that thei-e i^ "nothing in it." 



— The Meteor ologiselie Zeitsehrift for November contains a sum- 

 mary, by Dr. J. Hann, of the meteorological observations taken at 

 Cairo from 186S-88. The observations have been published in 

 extenso, together with a good introduction upon the climate, in 

 the Bulletin of the Egyptian Institute, and although similar obser- 

 vations have occasionally been published before, the present series 

 contains much new and useful material. The most striking feature 

 in the climate of this part of Egypt, as we learn from Nature, is 

 the Chamsin, the hot and dust-bearing wind which makes its ap- 

 pearance in March or April for about three to four days at a time, 

 and robs a large portion of the trees of their leaves. In the inter- 

 vals during which this wind is not blowing the weather is pleasant 

 and clear during spring-time, and the nights fresh and calm. 

 During the summer the north winds prevail, with high tempera- 

 ture, very clear air, and great dryness. Towards September hu- 

 midity appears with the rise of the Nile, the ground is at times 

 covered with heavy dew, and the heat becomes oppressive on ac- 

 count of the moisture. In October and November fog occasionally 

 occurs in the morning, and rain begins to fall. After this season 

 the temperature is uniform and pleasant. Snow is unknown, frost 

 very seldom occurs, and rain is not very frequent. The absolute 

 maximum temperature of the 21 years' period was 117° in August, 

 1881, which was also closely approached in May, 1880, viz., 116.4°. 

 The absohite minimum was 28,4° in February, 1880, and the mean 

 annual temperature was 70.5°. Rainfall is only given for the 

 years 18S7-S8, in which 87 and 1.67 inches fell respectively. The 

 relative humidity sinks at times even oti a daily average to 12 per 

 cent, and has been known to fall as low as 3 per cent at certain 

 hours. Thunder-storms and hail are vei-y rare. The original 

 work contains a long investigation on the connection between the 

 height of the Nile and the weather, a comparison bet« een the 

 present climate and that at the beginning of this century, and 

 several carefully prepared diagrams referring to all meteorological 

 elements. 



— At the monthly meeting of the Royal Meteorological Society, 

 Dec. 10, Mr. W. Marriott gave the results of the investigation un- 

 dertaken by the society into the thunder-storms of 1888 and 1889, 

 wh'ch he illustrated by a number of lantern slides. The investi- 

 gation was originally con6ned to the south-east of England, but 

 as this district was found to be too circumscribed, it became nec- 

 essary to include the whole of England and Wales. After de- 

 scribing the arrangements for collecting the observations and the 

 methods adopted for their discussion, Mr. Marriott gave statistics 

 showing the number of days on which thunder-storms occurred 

 at each station ; the number of days of thunder-storms in each 

 month for the whole country; the number of days on which it 

 was reported that damage or accidents from lightning occurred ; 

 and also the number of days on which hail accompanied the 

 thunder-storms. In 1888 there were 113 days and in 1889 123 

 days on which thunder-storms occurred in some part of the coun- 

 try. The number of days with damage by lightning was 33 in 

 1888 and 38 in 1889; and there were 56 days in each year on 

 which hail accompanied the thunder-storms. The tables of 

 hourly frequency show that thunder-storms are most frequent 

 between noon and 4 p.m , and least frequent between 1 a.m. and 

 7 A,M, Thunder-storms appear to travel at an average rate of 

 about 18 miles per hour in ill-defined low barometric pressure 

 systems, but at a higher rate in squally conditions. The author 

 is of opinion thit individual thunderstorms do not travel more 



than about 20 miles; and that they take the path of least resist- 

 ance, and are consequently most frequent on flat and low ground. 

 Detailed isobaric charts, with isobars for two-hundredths of an 

 inch were prepared for 9 a.m. and 9 p.m. each day for the month 

 of June, 1888. An examination of these charts showed that in- 

 stead of the pressure being so very ill-defined, as appeared on the 

 daily weather charts, there are frequently a number of small, but 

 distinct areas of low pressure, or cyclones, with regular wind cir- 

 culation; and that these small cyclones passed over the districts 

 from whieh thunder-storms were reported. Sometimes it is not 

 possible to make out well-formed areas of low pressure from two- 

 hundredths of an inch isobars, but there is a deflection of the wind 

 which shows that there is some disturbing cause ; and thunder- 

 storms have usually occurred in that immediate neighborhood. 

 The author b'^lieves that the thunder-storm formations are small 

 atmospheric whirls, in all respects like ordinary cyclones ; and 

 that the whirl may vary from 1 mile to 10 miles or more in diam- 

 eter. There are frequently several whirls near together, or fol- 

 lowing one another along the same track. The numerous oscilla- 

 tions in the barometric curve are evidently due to the passage of 

 a succession of atmospheric whirls; and it appears that lightning- 

 strokes are most frequent when these oscillations are numerous 

 Mr. F. J. Brodie read a paper " On the Prevalence of Fog in 

 London during the Twenty Years 1871 to 1890." The popular 

 notion that November is par excellence a month of fog is not con- 

 firmed by the figures given by the author. The number of fogs 

 in that month is, if anything, slightly less than in October or 

 January, and decidedly less than in December, the last-mentioned 

 month being certainly the worst of the whole year. The latter 

 part of the winter is not only less foggy than the earlier part, but 

 is clearer than the autumn months. In February the average ' 

 number of days with fog is only 6.6, as against 8.9 in January, 

 10.2 in December, 9 3 in October, and 8.8 in November. 



— A paper on " Siouan Onomatopes," by J. Owen Dorsey, was 

 read before the Anthropological Society of Washington, D.C., 

 Dec. 1, 1891. According to "The Century Dictionary," "an 

 onomatope is a word formed to resemble the sound made by the 

 thing signified." Mr. Dorsey finds in the Siouan languages many 

 onomatopetic roots, hence he suggests the modification of the 

 definition just given, making it read, "An onomatope is a word 

 or root formed to resemble the sound made by the thing signi- 

 fied " In the paper under consideration, the author gives examples 

 of onomatopes in seven languages of the Siouan or Dakotan family : 

 Dhegiha.Kwapa, Kansa, Osage, Tciwere, Winnebago, and Dakota, 

 all but the Dakota having been collected by himself since 1871. 

 In these languages, according to the author, there are sundry per- 

 mutations of sound, among which are sh and Icli, gh and z, dh and 

 n. The words in which these permutations occur are not always 

 synonyms ; but when we find a word in which, for example, sh is 

 used, we may safely infer that the language contains another 

 word differing from the former only in the substitution of kh for 

 sh, or that one language or dialect uses s/t where another employs 

 its coiTelative, kh. Most of the onomatopes found by the author 

 are dissyllabic, a few being monosyllabic and polysyllabic. Some 

 of the onomatopes were given with the notations of their respec- 

 tive sounds as they appear to the Indian ear; thus, the sound of 

 the plane and drawing-knife (s-s-s) becomes the root s'u; whence 

 the verbs, ba-s'u, to use a plane ; and dhi-s'u, to use a drawing- 

 knife. The sound of a waterfall, of sawing wood, etc., is kh + 

 (a prolonged sound), the onomatope being khii'-'e in Dhegiha, 

 khu'-wa-d'^^e in Kansa, khu'-we in Kansa and Osage, kho'kh'e in 

 Tciwere, and sho +kh in Winnebago (the o in the last being pro- 

 longed). The creaking of new shoes or the sound of fiddle-strings 

 (gi-gi-gi) evidently suggested the root gi'-ze ; whence ba-gi'ze, to 

 play a fiddle; and nan-gi'ze, to make (new shoes) creak by walking 

 (in them). Many other examples were given; but the reader 

 is referred to the American Anthropologist for January, 1892, for 

 the full article. 



— Among the recent appointments of Johns Hopkins graduates 

 are Alfred Bagby, Jun. (Ph.D., 1891), adjunct professor of an- 

 cient languages, South Carolina College ; Edward A. Bechtel 

 (A.B , 1888), professor of Latin, Yankton College, South Da- 



