April 26, 1877] 



NATURE 



557 





adjacent to, the place. Climates are, therefore, frequently 

 of a local nature, by which I mean of small superficial 

 extent. Thus many varieties of climate may coexist 

 about the same parallel of latitude, and even over a very 

 limited portion of that zone. Instead of saying, then, 

 that a whole country such as Britain has a certain mean 

 temperature, as ascertained by lumping together observa- 

 tions made at places of widely different character, level, 

 and exposure, we should rather say that there are in that 

 island mountainous districts with a certain mean tempe- 

 rature, districts of open plain, having another, and 

 sheltered districts and valleys having another ; while 

 parts near the sea-shore, have their own peculiar cha- 

 racteristics. To take a familiar case, we may refer to 

 the Isle of Wight, all parts of which, small though the 

 island be, can hardly have the same climate as Bonchurch 

 and Ventnor, which are the favourite retreats of invalids 

 in pursuit of health. That such local atmospheric distinc- 

 tions do really exist may at once be shown by a reference 

 to the varied distribution of plant life which, though no 

 doubt largely affected by the nature of the soil, is never- 

 theless to a considerable extent dependent on the exist- 

 ence of certain atmospheric conditions. 



If meteorological stations were to be established in 

 some place situate in a low latitude — such, for example, 

 as the Island of Java, we should be told, as I have more 

 than once been in similar cases, that though 4 feet above 

 the ground may be suitable for thermometers in Britain, 

 it would be quite preposterous for so hot a climate as 

 Java. Now if what were wanted was to ascertain the 

 amount of heat emitted directly by the sun, such a state- 

 ment might be correct ; for then the instruments should 

 be kept as clear of terrestrial influence as possible, and 

 by taking proper precautions we might perhaps make our 

 observations indifferently at sea or onland. But these would 

 not be observations of climate. Now, as in the case we 

 have supposed, it is the climates of Java and Britain that 

 are to be compared by ascertaining the amount of heat 

 communicated to thermometers by conduction and con- 

 vection of the air which has been heated by solar, and 

 cooled by terrestrial radiation, the observations must be 

 made on the islands themselves and not on the sea which 

 surrounds them, and by instruments placed at the same 

 level above the surface of the ground. It has been farther 

 objected that in very hot countries there are large districts 

 where canes or other kinds of jungle vegetation rise much 

 above the level of the thermometers, while in Britain there 

 is generally a grassy sward nearly 4 feet below them. 

 These differing kinds of vegetation nevertheless largely 

 influence the character of climates, and their effects ought 

 not to be eliminated even although it could be done. The 

 results which have been obtained in a jungle should not, 

 however, as I have already said, be mixed up with those 

 of other places which have a free exposure. The truth is 

 that by adopting different kinds of protecting boxes, and 

 by varying sufficiently their levels above the ground, we 

 may so far depress the temperature of a hot country and 

 exalt that of a colder, as instrunientally to equalise them. 



There is but one mode of getting results which shall be 

 comparable, and that is by adopting the same standard 

 height and the same standard form of protecting box. 

 The results may, however, be vitiated in another way by 

 placing the instruments near or under .shelter of buildings, 

 or still more, by the monstrous system of fixing them to 

 the walls of houses ; for masses of masonry or other build- 

 ing materials prevent either extreme from being recorded 

 by the instruments. It must also be kept in view that 

 however valuable continuous registrations may be, in 

 showing intermediate variations of temperature, no photo- 

 graphic self-registering thermometer hitherto constiucted 

 gives any result which can be regarded as correct because 

 it does not record the temperature of the air of the locality 

 and is not comparable with those of common thermo- 

 meters, nor even, perhaps, is ever comparable with those 



of other similar self-recording instruments. The house 

 or framework with which the instruments are necessarily 

 connected cannot fail variously to affect the mercury in the 

 bulb and thus to veil the results. The only mode of counter- 

 acting this influence is to have common thermometers in 

 the neighbourhood placed and protected in the usual way 

 and to record their indications eight or twelve times in 

 the course of the twenty-four hours. 



It must be kept in view that I have been speaking only 

 of local climates, or those which are subordinate to the 

 normal climate due to geographical position. That such 

 great climatic zones due to latitude exist and vary as we 

 recede from the equator towards the poles is abundantly 

 evident, both from the animal and the vegetable world. 

 The best mode of investigating these climatic zones 

 would be to select stations as little affected as possible by 

 surrounding vegetation, the instruments being exposed as 

 freely as possible all round and placed at the same level 

 above the ground, and as nearly as possible at the same 

 level above the sea, so as to avoid confusion with what 

 have been termed the climatic zones of altitude. For 

 this purpose I venture to suggest the use of an instrument 

 which I proposed in 1870,' the indications of which depend 

 on the heating up of a larj^e quantity of water or other 

 fluid contained in a thin glass globe which is freely ex- 

 posed to the sun's rays. When the water expands under 

 the influence of heat, the surplus fluid escapes into an 

 adjoining vessel in which it can be afterwards weighed. 

 On the other hand when the fluid is contracted by cold, 

 the deficiency is continuously supplied from a connecting 

 cistern kept always at the same, or sensibly the same, 

 level. By this automatic arrangement the whole of the 

 heat given out, however irregularly, by the sun, is con- 

 stantly treasured up. The readings of maximum and 

 minimum thermometers would also serve to correct errors 

 due to the proximity of the tubes and cisterns of the 

 instrument to which I have referred. The difference 

 between the results of this and the common thermometer 

 is the continuous registration of the alterations in bulk 

 produced by the variations of temperature ; whereas the 

 common thermometer fails to record the many changes 

 that take place between the maximum and minimum 

 readings, and which are due to sudden obscurations and 

 revelations of the sun caused by passing clouds during the 

 day, while the terrestrial radiation at night is similarly 

 affected. Even where this instrument is not used it would 

 I think be an improvement on the present system were 

 maximum and minimum thermometers kept constantly 

 immersed in a large globe of thin glass filled with water. 



Thomas Stevenson 



VOLCANIC PHENOMENA DURING 1875 



DR. GUSTAV TSCHERMAK'S Mineralogische Mittheil- 

 ungen (1876, 2) contain a most interesting accxjunt of the 

 volcanic occurrences during the year 1875, computed by Prof. 

 C. W. C. Fuchs. In the short introduction Dr. Fuchs expresses 

 his regret that the scientific academies and societies do not give 

 more general attention to this most important branch of geolo- 

 gical research, and points out that through the numerous and 

 universal relations of the institutions in question the statistics of 

 volcanic eruptions and earthquakes would become far more cor- 

 rect in details and numbers, than it is in his own power to make 

 them. The publication of the valuable information now given 

 by Prof. Fuchs therefore all the more deserves the highest praise 

 and attention. Dr. Fuchs divides the events into two classes, viz., 

 eruptions and earthquakes. The first volcano treated of is— 



Etna. — After the short eruption of August 29, 1874, which 

 lasted until the beginning of September, the mountain was per- 

 fectly at rest. Early in January* 1875, there were signs of new 

 activity in the shape ot repeated shocks, which, on the 8tb, 

 caused considerable damage near Acireale. But the shocks de- 

 creased again both in frequency and intensity, and a new period 

 of rest ensued until the beginning of October. At that time a 

 small crater on the south-side of the mountain became slightly 

 active. From December 19, smoke mixed with reddish vapours 

 ' Journ, Scott. Met. Soc, vol. iii. p. ii4- 



