April ii, 1901] 



NA TURE 



573 



amount of useful information, including the average mean tem- 

 perature and the mean maximum and minimum temperature at 

 Brussels (or Uccle) for each day of the year, and the monthly 

 means and extremes of all the principal meteorological elements 

 since 1833. There are several articles of special interest, 

 including one by M. J. Vincent on the history of meteorology 

 in Belgium from the earliest times until the establishment of 

 the Academy of Sciences at Brussels in 1773. The next 

 Anmtaire will continue the sketch from the creation of the 

 Academy until the foundation of the Royal Observatory in 

 1833. 



The Committee of the Bristol Museum record in their annual 

 report the extensive operations carried on at Brislington, near 

 Bristol, where the remains of a Roman villa have been found*. 

 The extensive foundation walls of the villa were laid bare, 

 showing the plan of its construction and many of its domestic 

 features. In addition to two fine mosaic pavements, a great 

 variety of relics of the Roman period were discovered. 



The " Physical Geography of the Texas Region," by Mr. 

 Robert T. Hill, forms Folio 3 of the Topographic Atlas of the 

 United States, issued by the U.S. Geological Survey (1900).^ 

 It is a finely-printed work, illustrated with numerous maps and 

 beautiful photographic representations of topographic forms, 

 mountains, plains and scarps, rivers and canyons. In the de- 

 scriptive text the author deals, for the most part, concisely with 

 his subject, showing, first of all, the relations between the geo- 

 logical formations and the scenery, and then describing the 

 principal features. 



The Maryland Geological Survey, which is under the direction 

 of Dr. Wm. Bullock Clark, has just commenced the issue of a 

 series of reports on the physical features of the counties of Mary- 

 land. The first report on Allegany County may be taken as a 

 sample of what can be done by an energetic and well-equipped 

 staff. It occupies 323 pages, is printed in remarkably clear 

 type, and is illustrated with numerous diagrams, pictorial views 

 and maps. Accompanying it is a folio atlas, with a colour- 

 printed geological map on a scale of i inch to a mile, and other 

 topographic maps. Many experts aid in the descriptions of various 

 subjects, such as physiography, stratigraphy, mineral resources, 

 soils, climate, hydrography, forestry, and fauna and flora. The 

 county, indeed, is one highly favoured, few regions being more 

 salubrious or more picturesque. Along its full length from east 

 to west the Potomac River meanders through a district of rich 

 farming lands and wild mountain scenery. Silurian, Devonian, 

 Carboniferous, Permian and Pleistocene formations are met with, 

 and the history of research among these strata is fully recorded. 

 The leading characters and many details concerning the forma- 

 tions are given, though with too scant particulars of the fossils 

 to please those who seek comparisons with equivalent strata 

 elsewhere. The rocks grouped as Permian follow the Carbon- 

 iferous conformably. They comprise shales and limestones with 

 unimportant sandstones and coal-beds, and their fossils have yet 

 to be described. To those residing in Allegany County this 

 admirable memoir cannot fail to be of the greatest interest and 

 service. 



Those who have visited the Bankfield Museum at Halifax 

 are aware of the improvements effected in the ethnographical 

 section by the untiring industry of Mr. H. Ling Roth, the 

 honorary curator, who has, in addition, just issued a pamphlet 

 on the Fijian Collection. This excellent little guide of twenty- 

 seven pages contains forty-four illustrations, mostly of specimens 

 in the collection. It forms an interesting sketch of Fijian eth- 

 nography, written with that carefulness of detail which students 

 have learnt to expect from Mr. Ling Roth. As coloured designs 

 on bark cloth are found in some parts of New Guinea where 

 NO. 1 64 I, VOL. 63] 



direct Polynesian influence is entirely out of question, there 

 seems no reason to believe, as Mr. Ling Roth suspects, that Fijian 

 inasi or tapa is a '* Polynesian institution introduced among this 

 Melanesian people." The same argument applies, though 

 perhaps not so conclusively, to tatuing. 



Where the Thompson and Fraser rivers meet at Lytton in 

 Southern British Columbia has always been an important site of 

 the Indians, as the ancient burial grounds and village sites testify. 

 The late Dr. G. M. Dawson first described the remains in 1891, 

 but recently the Jesup North Pacific Expedition made a series of 

 explorations in this vicinity, of which an abstract has been 

 described by Harlan J. Smith in the Memoirs of the American 

 Museum of Natural History, vol. ii. "^zxtmXMonuinental Records). 

 The prehistoric culture resembles that of the present inhabitants 

 of the interior of British Columbia. The mode of life of the 

 prehistoric tribes, their utensils and even their customs must 

 have been practically the same as those of recent Indians. There 

 are, however, a few slight differences ; the ancient type of pipe 

 resembles the prehistoric pipe of Oregon and California, while 

 the recent pipe is practically of the same type as that found on 

 the plains. The potter's art was then, as now, unknown. On 

 the whole the prehistoric culture of the interior of British 

 Columbia shows greater affinity to that of the western plateaus 

 than to that of the North Pacific coast. 



Mr. C. D. Child, writing in the Physical Review for 

 February, describes some experiments made with the new 

 method for determining the velocity of ions, recently suggested 

 by Prof. J. J. Thomson. The method in question "is to pro- 

 duce the ions in one region and measure the electrical inten- 

 sity at two points where there is no production of ions, but to 

 which ions of one sign only can penetrate under the action of 

 the electric field." The author shows by an application 

 of Prof. Thomson's method that the velocity of the positive ions 

 drawn from a Bunsen burner is approximately 2 '2 cm. per sec. 

 for potential gradient of i volt per cm., and that for negative 

 ions 2 "6 cm. Further, in the case of an unlimited supply of 

 ions, if the discharge takes place between two regular surfaces, 

 the velocity may be determined by simply measuring the current 

 per unit area and the difference of potential between these sur- 

 faces, and if the surfaces are not at all regular the relative 

 velocities of the positive and negative ions may be determined 

 by comparing the positive and negative currents. 



In a paper on stationary motions, published in the Atti dei 

 Lincei, x. 5, Signor T. Levi-Civita has endeavoured to furnish 

 a more precise definition than commonly exists of the conception 

 of stationary motion. Routh's definition, taken in its purely 

 formal aspect, leads to the conclusion that by a proper choice of 

 variables any motion may be regarded as stationary. On the 

 other hand, experience teaches us that certain motions possess 

 peculiar characters of simplicity and regularity which distinguish 

 them clearly from other motions, and, moreover, Routh's ex- 

 amples show that in certain cases his definition actually does 

 distinguish stationary motions (in the physical sense) from non- 

 stationary motions. The author considers that the distinguishing 

 characteristic in such cases is that the integrals or invariable 

 relations which determine stationary solutions are always 

 uniform in the sense considered by Poincare. According to 

 Routh, a stationary motion 2 is characterised by the property 

 that if the conditions are equally modified at any two instants, 

 /', t"f the disturbed motions 2', 2" present relations such that 

 under a certain condition they may be regarded as equivalent. 

 Now Signor Levi-Civita considers that an analytical condition 

 which is not uniform has no physical interest, and he is of 

 opinion that Routh's definition of stationarity should be com- 

 pleted by adding the proviso that the relations between two 

 disturbed motions, 2', 2", should be uniform. As an example. 



