January -8, 1903 J 



NA TURE 



255 



not appear that mere isolation suffices to produce even distinct 

 varieties of Calochortus. For ins'ance, C. catalinae^ Watson, is 

 found on Catalina and other islands, and also on the mainland ; 

 but instead of running into numerous insular races, it " is one of 

 the least variable " of all, and no variety has been distinguished 

 by name. Oh p. 141, Mr, Purdy admits that his Calochortus 

 I'cniulus, var. eldorado, " var. nov.," is the same as C. venustus 

 purpurascctts, Watson ; while he applies the name pwpurascen s 

 (Purdy, 1895) to a quite different variety of the coast range. 

 This surely cannot be permitted ; the former must stand as 

 durpurascens, while the latter may be called var. Caroli. 



T. D. A. COCKERELL. 



EARTHQUAKE OBSERVATIONS IN GALICIA. 



T"HE ninth number in the new series of the publications 

 -*- of the Austrian Academy of Sciences relates to 

 earthquakes observed during the year 1901 in Lemberg. 

 The first feature which one observes in this publication, the 

 author of which is Dr. W. Laska, is that he describes each 

 earthquake according to the phases it exhibits, the various 

 phases being distinguished from each other by differences in 

 their periods. Twenty years ago, earthquakes were described 

 as consisting of preliminary tremors, shocks and concluding 

 vibrations, each of which had distinguishing periodic motions. 

 Now we find first preliminary tremors of types // and p r ", 

 second preliminary tremors of types p s ', p 2 ", p<!" and /■■>"", and 

 on they go, commencing with //, with periods between 2'i and 

 69 seconds, and ending with types where the periods have 

 exceeded one minute. Inasmuch as these groups overlap, so 

 that it is frequently difficult to assign a set of waves to their 

 proper group, for our own part we are for the present content 

 to divide the seismic spectrum into four parts — first and second 

 preliminary tremors, large waves and concluding vibrations. 

 In addition to these entries, Dr. Laska gives tables of tri- 

 daily readings of two levels and of a thermometer. The most 

 interesting portion of the work is, however, found in its intro- 

 duction, where, amongst other matters, reference is made to the 

 natural period of a pendulum as influencing the magnitude of 

 its records and to rules which enable an observer to determine 

 the distance of an origin from the inspection of a seismogram. 



One simple rule is to diminish the duration of the first pre- 

 liminary tremors reckoned in minutes by unity and multiply 

 the same by 1000. The result is an approximation to the 

 distance of the origin expressed in kilometres. For example, 

 if a seismogram shows that the preliminary tremors had a 

 duration of 7 '6 minutes, then the earthquake it represents 

 originated at some place about 6600 kilometres distant. The 

 mnemonic is certainly simple, but its application is con- 

 fined to those records where preliminary tremors are well 

 defined. These are comparatively few in number and the 

 accuracy of the determination is dependent upon the measure- 

 ment of intervals of time which are small. These objections 

 apply to a second rule suggested by Dr. Laska, the value of 

 which is apparently still further impaired by the introduction of 

 two assumed constants determined by Dr. F. Omori. These 

 constants are the velocities of the first and second preliminary 

 tremors as determined from observations of ten earthquakes 

 which originated near Japan and were recorded at Tokio and 

 in Italy. 1 To obtain these velocities, the arcual distance be- 

 tween the Tokio isoseist and Italy is divided by the difference 

 between the times of observation in Tokio and Italy. Had the 

 distance between the origins and Italy been divided by the 

 difference of times between the times of origin (which are easily 

 calculable) and the times of arrival in Italy, then the constants 

 given by Dr. Omori would have been reduced. A further 

 reduction would be made on the assumption that the wave paths 

 of the motion considered had approximated to chords. If the 

 speed of the preliminary tremors between their origin to the 

 Tokio isoseist had been the same as it was from that isoseist to 

 Italy, then the above objections might be withdrawn, but this, 

 according to Dr. Omori's own showing, appears hardly to be 

 the case. 2 



Although it is interesting to find the relationship between the 

 duration of preliminary tremors and the distance they have 

 travelled again brought to our notice, the well-known method 



1 " Publications of the Earthquake Investigation Committee in Foreign 

 Languages," No. 5, pp. 71-80. 1. Tokio, 1901.) 



2 Jour. Sc. Coll., Tokio, vol. xi , p. 158. 



NO. 1732, VOL. 67] 



of determining origins by the interval of time between the first 

 motion of an earthquake and the subsequent arrival of the large 

 waves is apparently one of more frequent and certain applica- 

 tion. 1 J. Milne, 



PILOT CHARTS OF THE METEOROLOGICAL 

 OFFICE. 



1 N addition to the usual information, the Meteorological Offi 'e 

 pilot chart of the North Atlantic and Mediterranean for the 

 month of January deals with some new features, necessitating the 

 use of the back of the chart as well as the front. There is an 

 account of the destructive cyclone which visited our coasts on 

 October 15-16 last, and also of the slow-moving disturbance 

 and its accompanying gales which wandered about the 

 Tyrrhenian Sea from October 22-29. A summary is given of 

 the characteristics of the surface temperature of the Atlantic 

 for each of the ten months from January to October last, the 

 most striking feature being the evidence of a distinct tendency 

 for the water in the immediate vicinity of western Europe to 

 remain cooler than the normal during the first nine months, a 

 fact which may be associated with the persistent low air 

 temperature over the adjacent land during the spring and 

 summer. On the Newfoundland banks, there was a marked 

 excess of warmth through the first six months, little or no ice 

 being found in the locality. In October, an excess was shown on 

 the eastern side of the ocean for the first time, and simultaneously 

 the air temperature over the British Isles ■ passed above the 

 average in all districts. With the object of discovering what 

 connection, if any, there is between the movements of weather 

 systems and the distribution" of the temperature of the surface 

 water, observations are being collected for obtaining the n.ean 

 barometric pressure month by month over the region from 30° to 

 60° N., o" to 70 W. , and the tracks of the centres of storm 

 areas. For October, the mean isobars are superimposed on the 

 sea temperature results, while the storm tracks are given on a 

 separate chart. 



To arrive at any definite conclusion as to cause and effect, it 

 will require a long series of such charts — probably, too, for 

 shorter periods than a calendar month, periods determined by 

 the prevailing type of conditions, depending mainly on the 

 positions and stability of the controlling anticyclones. Sum- 

 maries are given of the ice reports from the whaling steamer 

 Balaena, up Davis Strait, and the barque Lady Head, in 

 Hudson Bay, last summer. Neither vessel passed any ice in 

 the lower part of Davis Strait when heading for home in 

 October. On July I last, the New Zealand Shipping Company's 

 s.s. Waikato was disabled in 33° S. , 6° E., and for twenty-six 

 days she drifted helplessly about the south Atlantic, being finally 

 taken in tow on July 27 in 28" S., 13° E., having in the interval 

 travelled S12 miles, or at an average rate of more than thirty - 

 one miles per day. The track of her wanderings day by 

 day, together with the direction and force of the wind, sup- 

 plied by Captain Kiddle, is reproduced, with the addition of 

 the normal current circulation of the region, which shows that 

 the Waikato followed closely the drift indicated by the Admiralty 

 chart. 



STARVING A PARASITE. 



TN a recent paper read before the Royal Society, 2 Prof. Marshall 

 Ward described the results of three series of experimental 

 cultures of Brome-seedlings in sand, to which had been added 

 various nutritive salts, or manurial mixtures, which were then 

 infected with the parasite to see how the latter behaved on 

 starved seedlings. Some of the seedlings received all the salts 

 necessary for successful development, others none of such salts 

 other than the root-hairs could extract from the sand itself and 

 from the reserves in the endosperm, and others all necessary 

 minerals except phosphorus, or potassium, or magnesium, or 

 calcium, or nitrogen respectively. 



So far as the seedlings themselves are concerned, the effects of 

 the mineral starvation were most evident in the small stature, 



1 "Brit. Assoc. Reports," 1900, p. 79; and " Seismological Investiga- 

 tion Report," 1902. 



2 " Experiments on the Effect of Mineral Starvation on the Parasitism of 

 the Uredine Fungus, Puccinia dispersa, on Species of Bromus." By Prof. 

 H. Marshall Ward, F.R.S. Read before the Royal Society on November 

 27. 



