August lo, 1905] 



NATURE 



34: 



fact that such tender plants as Cistus purpureus, Lam., 

 Chciranthus mutabilis, L'H^rit., and many others have 

 survived the last six winters unprotected ; while large 

 bushes of Laurustinus, Euonyinus japonicus, bay, &c., 

 were evidently little, if at all, injured by the terrible 

 winter of 1895. Yet, even within the limits of my own 

 grounds, with a rise of only 35 feet up to the 400-feet 

 contour line, there is a marked difference of climate. On 

 November 27, 1004, in the upper part of the garden, 

 dahlias planted within 4 feet of a high wall facing south 

 were blackened by frost, while in the lower garden those 

 in the open border were uninjured. 



The difference between the climate of this place and 

 the Public Gardens at Maidstone is fairly shown by the 

 following comparative statement, the temperatures from 

 the latter having been taken w-hen I chanced to pass the 

 place where they are put up, and therefore not selected : — 

 Max. in Min in Min. on Range 



M?y II, 1904 shade shade grass in shade 



Ulcombe 50-0 



Maidstone ... 56 'o 

 May 28 to 30. 1005 

 Ulcombe, 2Sth 676 



,, 29th 70-6 



30th 760 



Maidstone, 2Sth 76-0 



., 2gth 810 



,, 30lh 85 'o 



The maxima in both cases are those of the previous 

 day. Maidstone is seven miles from here, and lies in 

 the valley of the Medway. 



Vet, in spite of the fact that the thermometer, even on 

 the grass, has not been below 32° since April 3,' we are 

 no better off for apples than our neighbours ! The apples 

 did not begin to flower until the end of April, so some 

 other cause than frost must be found to account for the 

 bad crop. This is an e.xample of the difliculties of the 

 question ; other complications are the nature, mechanical 

 and chemical, of the soil ; period of blooming of 

 different varieties of the same fruit ; shelter from the 

 generally prevailing cold winds in spring, &c. Still, some 

 effort should be made to ascertain the conditions under 

 which, on an average of years, the best crops can be 

 obtained, and so avoid the waste of time, money, and land 

 that has been incurred in hundreds of instances by plant- 

 ing orchards in unsuitable localities, while hundreds of 

 acres of suitable land are used for corn and other crops 

 that would grow as well elsewhere. 



Alfred O. AWalker. 

 Llcombe Place, near Maidstone. 



Islands for Weather Forecasting Purposes. 



In Nature for June i is a very suggestive article by 

 Dr. Kockyer under the above heading, in which specific 

 reference is made to the meteorology of Western Australia. 

 It is becoming increasingly evident that the Indian Ocean 

 and its neighbouring continents form one of the most 

 interesting fields in the world for the study of meteor- 

 ology, and as the officer-in-charge of an important section 

 of this region I am most anxious to assist in this study 

 in any way possible. Our progress will be slow if we 

 start WNth incorrect theories, and my present object is to 

 point out the probable inaccuracy of a few of the funda- 

 mental concepts, and to indicate briefly a few of the 

 observed facts which seem to have a bearing upon the 

 whole matter. 



There is little or no rain in Perth of a monsoonal 

 character. The wettest months are May, June, Julv, and 

 August, during which time the prevailing winds are not 

 from the S. or S.W. Rain is almost alwavs associated 

 with the passage of a " low " along the south coast, 

 setting in with the wind at N. or N.W., and fi)iisliiiig 

 when the wind veers to .S.W. and S. 



There is a tendency throughout the year for the winds 

 to alternate from the eastward during the forenoon to the 

 S. or S.W. in the afternoon. This is most marked in the 

 summer months, when the prevailing feature of the weather 



1 Vet severe frost with great damage ti crops in the Dartford. Rocheste-, 

 and Hot districts; ako at Maidstone and Sevenjaks on May 22-3, is 



NO. 1867, VOL. 72] 



map is a " high " stretching along the ocean south of 

 our coastline. How far south or west this extends I 

 cannot sav. The prevalence of southerly winds in the 

 summer time is probably due to this anticyclonic area, 

 and Fig. 2 on p. 11 1 is therefore soinewhat misleading. 



As the sun moves north the high pressure follows it, 

 and in June and July forms a belt across the centre of 

 Australia. It is, however, constantly on the move from 

 west to east. A " high " will generally during these 

 months strike the west coast about, or to the north of, 

 Perth, and gradually work across to the eastern States. 

 As it passes our wiiid sets in strongly from the eastward, 

 gradually veering more northerly. By the time the 

 " high "' reaches, say, Adelaide, our wind is N.N.E., the 

 isobars are running nearly parallel to the west coast, and 

 we are looking out for a "low" to approach from the 

 ocean. As a general rule, the " low " is first heralded from 

 Cape Leeuwin, the extreme S.W. corner of Australia, but 

 rain sets in with a N. and N.W. wind all along our west 

 coast as far as the N.W. cape. It is heaviest in the 

 extreme S.W. The " low " generally passes south of 

 Cape Leeuwin and across the bight to Tasmania. So 

 long as our wind, and especially that at the Leeuwin, has 

 any northerh component, we are pretty certain to have 

 inore rain, bijt as soon as it reaches W.S.W., and especially 

 S.W., we anticipate clearing weather. 



Whence these " lows " come before they reach us is 

 therefore a question of great importance. I believe the 

 usual theory upon this point is incorrect. That is, that 

 these " low's " are northerly extensions of the Antarctic 

 low-pressure belt, which sweep past the Cape of Good 

 Hope, and after the lapse of a few days reach Cape 

 Leeuwin, and so travel along the south coast of Australia. 

 I think this is incorrect for several reasons. In the first 

 place, I have endeavoured to trace notable storms either 

 forward from the Cape to Australia, or backwards from 

 Australia to the Cape, and have not been able to find 

 any connection whatever. Secondly, from theoretical con- 

 siderations, a rotating body of air in the latitude of the 

 Cape would possess a sufficient southerly component to 

 its motion of translation to carry it well south of Australia. 

 Thirdly, the inore direct evidence stated in the next 

 paragraphs. 



During the summer months, January, February, and 

 March, there is a class of storm which strikes our N.W. 

 coast and then travels across the State in a S. or S.S.E. 

 direction, emerging in the Great Australian Bight, and 

 travelling thence in an E.S.E. or S.E. direction towards 

 Tasmania. Before striking the N.W. coast it can sotrie- 

 times be traced from the extreine north of the State moving 

 towards the S.W., down the coast, but keeping well out 

 to sea, then gradually recurving, and striking the coast 

 about lat. 20°. The existence of this class of storm and 

 its approximate path is now beyond doubt, though until 

 recently it was ignored in practical Australian meteor- 

 ology. I think, however, it would now be safe to say 

 that it dominates the weather of at least the western and 

 southern portions of Australia during the summer months, 

 though on account of the paucity of stations in its track 

 our knowledge of the various conditions is at present 

 elementary. It is important to bear in mind that the study 

 of Western .Australian meteorology is in its infancy. Not 

 until the last few years was the iinportance of this class 

 of disturbance recognised, and therefore any theories which 

 had been formed require to be modified. During the last 

 two years evidence seems to me to be accumulating that 

 this particular class of storm persists throughout the year, 

 and is, in fact, the dominating influence in Australian 

 meteorology. If this be so, it can easily be seen how 

 profoundly' older theories are afTected, and how necessary 

 it becomes to make a fresh start. 



Even during the summer the disturbances do not all 

 follow along the same track. Sometimes they strike the 

 coast near or even south of the N.W. cape, and 

 occasionally they just miss the coast, but can be traced, 

 following it down, but keeping out to sea, and eventually 

 rounding Cape Leeuwin and behaving like an ordinary 

 winter storm. It is this latter path to which I wish to 

 direct special attention. 



In the winter, as a general rule, the first intimation of 

 an approaching " low " is obtained from Cape Leeuwin, 



