conclusion rather more than a year ago. In addition to the results of the Challenger observations, he also made use of records of temperature, atmospheric pressure, etc., received from a large number of stations in all parts of the world. Some of the most striking points in the report are given in an address to the Royal Geographical Society, published in the Proceedings for March and accompanied by four maps, of which two show the distribution of temperature and atmospheric pressure, respectively, for the month of January, and the other two the same phenomena for July. These are reproductions of some of the fifty-two maps annexed to the report. One important fact that the Challenger observations revealed is, that the daily variation of the temperature on the surface of the ocean away from land is very small, nowhere exceeding a degree between lat. 40 N. and 40 S., and falling to onefifth of a degree in high latitudes. The temperature of the air was found to have a range about three to four times as great as that of the water below. In the Southern Ocean, at about lat. 63°, it was 0°8 inches, or four times as great as that of the sea in the same region. Over the open sea the humidity curve closely follows that of the temperature, falling to a minimum at four o'clock in the morning and rising to a maximum at two in the afternoon. But near land a second minimum occurs from about 10 A.M. to 2 P.M. At this time, the land being heated, a current rushes in from the sea to take the place of the hot air that rises from it, and dry air from the upper regions of the atmosphere descends over the ocean. Over the open sea the barometer, though removed from the disturbing influence of land, shows as marked oscillations as over land where the diurnal variation of temperature is great. The cause must be sought in the daily changes in the temperature and humidity of the air produced through all its height by solar and terrestrial radiation. Another important fact is that, latitude for latitude, the amplitude of the barometric oscillations is larger in an atmosphere highly charged with aqueous vapour than in a dry one. In the anticyclonic regions of the Atlantic and Pacific, the barometer falls only about 0·025 inches from the morning maximum to the afternoon minimum. Since pressure remains high, though currents of air are constantly flowing out from these regions in all directions over the surface of the ocean, it follows that the dry air from above must descend into their centres. These anticyclonic regions play a most important part in regulating the climates of the neighbouring continents. The four principal lie in the Atlantic and Pacific, at about lat. 36° N. and S., and appear in all the monthly charts, with the exception of the North Atlantic region, which is absent in the month of January only. The absolutely highest mean pressure for any month, about 30-5 inches, is to be found in Central Asia in the month of January. Here, to the south of Lake Baikal, is the centre of a great anticyclone, covering a large part of Eurasia, from which south and south-west winds blow over Russia and Western Siberia, raising the temperature of these countries. Their effect may be seen on the temperature chart, on which the isothermals run nearly north and south. Another example of the effect of pressure on climate may be taken from the low-pressure system in the North Atlantic, where the lowest mean pressure of 29.5 inches occurs between Iceland and the south of Greenland. This system gives rise in winter to south-westerly winds in Western Europe, and northwesterly over North America. While, therefore, the temperature of the former is abnormally raised by winds from lower latitudes, that of the latter is lowered by cold breezes from the Arctic regions. Hence, the temperature of coast of Labrador is only - 13°, while on the same parallel in Mid-Atlantic it is 45°, or 58° higher. The influence of other cyclonic and anticyclonic areas is discussed in Dr. Buchan's article. In reference to the drawing of isobars, the author gives a warning against the use of observations in steep and confined valleys, where descending cold currents at night and ascending warm currents in the afternoon unduly raise and depress the barometer alternately. Thus, in the Valley of Tönset, in Norway, the mean is 29.95 inches, while at Dovrè, situated at about the same elevation but separated from Tönset by a broad range of mountains, it is 29.87 inches. Lastly, a few figures must be quoted regarding the velocity of the wind. This the Challenger observations showed to be greater over the open sea than near land, the mean difference being from four to five miles per hour. It is greatest over the Southern Ocean (23 miles per hour) and least over the North Pacific (15 miles). The curves on the open sea show a very slight diurnal variation, but near land they exhibit a distinct minimum between 2 and 4 A.M. and a maximum from noon to 4 P.M. The difference between the velocities on sea and land is greatest at 4 A.M., and gradually falls to a minimum at 2 P.M., demonstrating the effect of the land in reducing the velocity by friction and the fact that this effect is, in some way or other, partially counteracted by the heating of the surface of the land. Such are a few of the important results pointed out in Dr. Buchan's paper, which is so full of valuable information that no abstract can do it justice. The Mean Heights of Continents.-Many calculations have been made of the areas of continents and oceans and their mean heights and depths, respectively. Within the last ten years De Lapparent, Murray, Penck, Supan, and Tillo have devoted much time and care to the working out of tables of mean heights. De Lapparent's results, however, though they exhibit no striking divergence when compared with the others, cannot be relied on, owing to his arbitrary assumption of mean heights for the zones bounded by his contour lines, while Penck's and Supan's tables are deduced from Murray's data, and Tillo made his planimetric measurements on the maps prepared for Dr. Murray by Mr. J. G. Bartholomew. A new and independent solution of the problem is, therefore, desirable, and Dr. Heiderich's treatise, in Dr. Penck's Geographische Abhandlungen, bd. v., heft 1, merits the attention of all who take an interest in the problem. Dr. Heiderich has not only collected his own materials, but has calculated his areas and mean heights by a new method, namely, by the use of Simpson's formula. He ascertains, first of all, the extension of the land along the equator and along parallels of latitude at intervals of five degrees, leaving out the almost unknown areas to the north of 80° N. latitude and to the south of 70° S. latitude. To calculate the volumes of land above the sea-level Dr. Heiderich has drawn sections along each of the abovementioned parallels, of which reduced copies are annexed to the pamphlet. The area of each section is calculated, and three adjacent ones are combined to find the volume by Simpson's formula. It is interesting to consider what probability of error is involved in the use of this formula If y1, y2, and y, be the lengths of three parallel lines drawn at equal intervals across any surface, its area is X (Y1 × 4y2+Y3), where h is the distance between the lines. But the result is exact only when y = a + bx + cx2 + dx3, that is, when y1, y2, and y, are equal to ordinates of one of the straight lines and parabolic curves represented by the equation. The accuracy of the area given by the formula will, then, depend on the position of the central ordinate. Should it happen to terminate in a gulf or peninsula, the area will probably be too small in the first case and too large in the second. As, however, peninsulas and gulfs usually run north and south, and the vagaries of the coast-line between ordinates 5° apart will be small compared to the longitudinal extension of the land, it is not surprising that Dr. Heiderich's calculation of the land area between 80° N. and 70° S. latitude is only one-twentieth per cent. in excess of Dr. Wagner's, which was obtained by the planimeter. In individual h 3 continents the errors are likely to be greater, and there is a smaller probability of their counterbalancing one another. Dr. Heiderich has already made a planimetric calculation of the area and mean height ef Africa, and he compares the results. Among the errors of his present method he points out that this continent maintains a considerable width to the south of 30° S. latitude, and approaches close to the 35th parallel without reaching it. Consequently the area of the zone between the parallels 30° and 40° is only 94 square miles as given by the formula, whereas the planimeter gives 181 square miles. There is a still greater probability of error in the calculation of the vertical sections; and here again Dr. Heiderich points out a failure of a similar kind to that just explained. There is no section passing through the highlands of Abyssinia. The net result of these and other errors is that he obtains 1975 feet for the mean elevation of Africa instead of the 2198 feet given by the planimetric measurement. Dr. Heiderich's mean heights are as follows: The first point that strikes one is the great elevation of Europe, which Dr. Heiderich attributes to his having included Iceland, Spitzbergen, Novaia Zemlia, and Trans-Caucasia. His height, however, exceeds Dr. Murray's by 291 feet, and Tillo's by 190 feet, and a rough calculation shows that a layer 190 feet deep over Europe, Trans-Caucasia, and the islands mentioned would raise the average height of the latter to about 5900 feet, while, if Dr. Murray's figures be taken, the average height would be over 7500 feet. Now, the loftiest of the islands, Novaia Zemlia, has a mean height of only 2130 feet, and, lofty as Trans-Caucasia may be, it can hardly supply these deficiencies. We must, therefore conclude that Dr. Heiderich's estimate for Europe is considerably in excess even of Tillo's. Very high figures are also given for North America and Australia (including New Guinea), and though the explorations of Nansen and Sir W. Macgregor warrant the assumption of greater mean heights for these lands than have been hitherto assigned to them, we are inclined to think that, in the former case at any rate, the height given by Dr. Heiderich is excessive. Dr. Heiderich has collected a large number of data, and taken considerable trouble in working them out, but, considering the defects he has himself pointed out and the probability of the existence of many others, it would be rash to affirm that his calculations merit greater confidence than those of his predecessors. If Simpson's formula be used, regions where it fails should be carefully searched for, and errors rectified by drawing additional sections, or by some other method. Even then, probably, a more accurate solution of the problem may be expected from increasing the number of contour lines and using the planimeter. Trade-Routes between the Cape of Good Hope and Australia.-It may be seen by laying a ruler on a gnomonic projection chart that the shortest route between Cape Agulhas and Cape Leeuwin nearly touches the 46th parallel of S. latitude, while between Cape Agulhas and Melbourne it cuts the 57th parallel. But, when the ice in high latitudes and winds are taken into consideration, the choice of the best route becomes more difficult. Captain M. W. Campbell Hepworth, in a paper read before the Royal Meteorological Society (Quarterly Journal, January 1891), discusses this question. Lieutenant Maury advocated the higher parallels, not simply because the distance along them is shorter, but because, in his opinion, the winds and seas south of 40° are more favourable; but many authorities consider that the evidence in support of this opinion is by no means conclusive. Captain Hepworth, however, after inspecting the log-books of several steamers of the Anglo-Australasian line, has decided in favour of a route between the 41st and 42nd parallels during the winter, and between the 45th and 46th in summer. The steady flow of westerly winds in the South Indian Ocean is interrupted, occasionally in summer and frequently in winter, by gales of cyclonic origin, which, east of the 30th meridian, seem to travel on paths south of the 43rd parallel in the latter season, and south of the 46th parallel in the former. Therefore, by keeping somewhat northward of the 42nd parallel in winter, and of the 46th in summer, the navigator would make the utmost use of the strong fair gales which blow on the outskirts of these cyclonic areas, while by adopting a more southern route he might approach too close to the centre of depression, or encounter the adverse winds on the right of the centre. As regards ice, none was seen on thirteen out of fourteen voyages on which the easting was made south of 40° S. latitude, and on many of these the 44th parallel was crossed. MISCELLANEOUS. The first section of the Jaffa and Jerusalem Railway, namely, up to Er Ramleh, was opened for traffic on 24th May 1891.-The Times, May 30. Dr. Ten Kate has been commissioned by the Amsterdam Geographical Society to explore the smaller Sunda Islands. He will visit Sumbawa first, and then proceed to Flores.-Petermann's Mitt., No. 4, 1891. A New French Expedition to Lake Tsad has been equipped. It is under the command of M. Jean Dybowski, a lecturer at the Agricultural College of Grignon. By the end of May it had probably already commenced its march from Loango.The Times, May 23. Early in May an expedition under Mr. R. E. Peary was to leave New York for the Arctic Regions. At Newfoundland the party was to take passage in a sealer to Whale Sound, whence they hoped to reach the northern point of Greenland by land. Goldthwaite's Geog. Mag., May 1891. M. Joseph Martin, whose expedition was referred to in vol. vi. p. 49 of this Magazine, has travelled from Pekin through Lan-chow to Su-chow, where he met the brothers Grum Grzhimailo. He had experienced great opposition from the Chinese officials, and was in very bad health.-Compte Rendu of the Paris Geog. Soc., No. 5, 1891. The High Commissioner has proclaimed the Queen's sovereignty over the strips of territory known as the Bastard's Country, lying to the west of the crown colony of Bechuanaland. The country extends from the Orange River to the Nosob River, and is bounded on the west by the 20th meridian of E. longitude, the frontier of Namaqualand. This summer Dr. Joseph von Siemiradski of the University of Lemberg is to lead a Scientific Expedition to South America. After examining the lands in South Brazil which are suitable for colonisation, the expedition will cross the Pampas with the object of investigating the geography, geology, and fauna of the basins of the Rio Negro and Rio Colorado as far as Lake Nahuelhuapi and the Chilian frontier.-Deutsche Rundschau, Jahr. xiii. No. 6. 387 The use of Granite in the United States has largely increased during late years, and the rock has now assumed an important position among the mineral products of that region. From a report by Mr. R. B. Porter it appears that the value of the granite produced in the United States during the year 1889 was 14,464,095 dollars, as against 5,188,998 dollars in 1880. The four most productive States at present are Massachusetts, Maine, California, and Connecticut, in the order named, but those showing the greatest percentage of increase over the returns for the year 1880 are Minnesota, New York, Delaware, and Georgia. On May 2d the Canadian Pacific Express reached Montreal, having crossed the continent from Vancouver, a distance of 2900 miles, in the remarkably short time of 91 hours. The passengers reached Vancouver by the Empress of India, which made the passage from Hong Kong in 514 hours 23 minutes, stops included. The letters brought by this route from Shanghai were thirty-two days in transit to London, and those from Yokohama twenty-five days, and they might have been delivered three days earlier had there been a steamer at New York ready to take them on at once.-Sir G. Baden Powell in The Times, 14th May 1891. We have received the prospectus of the fifth session of Vacation Art and Science Courses (of which Professor Patrick Geddes is the moving spirit), to be held in Edinburgh in August next. In the school of Social Science a course of lectures will be given by Professor Geddes, Dr. Ernest Grosse of Freiburg, and Professor A. C. Hadon. In connection with this theoretical course there will be a course of practical studies in Seminar, conducted by Professor Geddes, Mr. A. J. Herbertson, and Mr. A. von Jankovich; and a course of studies in Studio, directed by Miss M. R. Hill-Burton, and Miss Alice Gray. These theoretical and practical courses will be supplemented by the production of Tableaux Vivants, and by excursions in Edinburgh, and to the Kingdom of Fife, the Clyde, and the Border. In the school of Natural Science there will be a course of General Biology by Professor Geddes and Mr. J. Arthur Thomson; a course of Zoology, conducted at Granton Marine Station by Mr. Arthur Thomson, assisted by Mr. Norman Wyld and others; and a course of Botany by Professor Geddes and Mr. R. Turnbull. The general aim of these courses is to substitute demonstrations on the actual objects studied for formal lectures, and at the same time to get rid of the artificial stimulus of examination and competition, and to make education a preparation for life. We wish these courses all success. NEW BOOKS. Canada and the Canadian Question. By GOLDWIN SMITH, D.C.L. With Map. The author of this work has so long resided in Canada, and is so well known as a thinker and writer, that importance will naturally attach to his latest publication. He commences by showing that the Dominion of Canada is divided into four widely-separated "blocks," connected only by railways, viz. :-1, The Maritime Provinces (Nova Scotia, New Brunswick, and Prince Edward Island); 2, Old Canada (Quebec and Ontario); 3, Manitoba and the North-West Provinces; and 4, British Columbia. He considers that each of these "blocks" has more affinity with the adjoining United States' territory than with the other "blocks," from which it is divided by physical barriers. He then reviews in order the various divisions of Canada, beginning with the "French Province" of Quebec. He tells |