488 



NA TURE 



[March 22, 1906 



at the third station only managed to obtain some 

 results at third contact, the weather being very un- 

 favourable. 



At Fort de Kock, in addition to some beautiful 

 corona pictures (one of which is here reproduced 

 for comparison in Fig-, i) taken with the 40-feet 

 coronagraph, used horizontally in this eclipse, photo- 

 graphs were secured with the 30-feet concave grating. 

 In the latter, films were used in consequence of the 

 sharp curve in the focal plane of the grating, but, 

 as the report says, " each film showed that, un- 

 fortunately, it had not been placed exactly in focus, 

 still the dispersion was so great that many of the lines 

 could be very easily identified." A table is given 

 showing the results obtained from the measures of 

 these negatives, the spectrum covering A 3118-5 to 

 x 5 2 °4'7 ; intensities, character, and wave-lengths from 

 Rowland's tables are also included. 



At Sawah Loento the plane grating proved a 

 success, parallel rays falling on its surface and being 

 brought to a focus on the photographic plate by means 

 of a lens placed between the grating and the plate. 

 In spite of clouds, the negative taken at third contact 

 is said to have been fully exposed. The large dis- 

 persion employed and the definition obtained allowed 

 wry accurate wave-lengths to be deduced, so that the 

 table of wave-lengths extending from A 3835.2 to 

 x 4957-8 will be very valuable to compare with those 

 made by other observers. 



The "discussion of these results is here carried to 

 some length, but space does not permit of any 

 extensive reference. It may, in the first place, be 

 said that both Mr. Jewell and Dr. S. A. Mitchell 

 record having observed the magnesium (A. 4481) line 

 in the photograph of the chromospheric spectrum, and 

 both agree in the determination of the wave-length, 

 intensity, and length of arc. It is described as being 

 stronger in the chromospheric spectrum than in the 

 ordinary solar spectrum. The presence or absence of 

 this line in the chromospheric spectrum is a point of 

 such great importance that the observation above de- 

 scribed requires to be very carefully corroborated before 

 it can be finally accepted. It is, however, very difficult 

 to understand how the above identification of the 

 magnesium line with the chromospheric line has been 

 obtained, because in the list of wave-lengths here pub- 

 lished the evidence seems to point to a titanium origin. 

 Thus we find in this table that the wave-length of 

 the chromospheric line, as measured, is A 448L4, 

 while the solar lines nearest this are, according to 

 Rowland. A 4481.298 (Mg) and a 4481.438 (Ti). 

 Further, has it been definitely established that the 

 solar line A 4481.298 is due to magnesium? 



It is also stated that it seems probable that the more 

 volatile gases of atmospheric air uncondensed at the 

 temperature of liquid hydrogen, together with 

 hydrogen, helium, neon, and argon, are present in 

 the chromosphere, but with regard to krypton and 

 xenon the evidence is not conclusive. These deduc- 

 tions also do not seem to be supported by sufficient 

 evidence, but will require further discussion before 

 they can be generally accepted. 



Enough, perhaps, has been said to indicate to the 

 reader the importance to the studv of solar physics 

 of the publication of such a volume as this. Here 

 "' have all the data and discussions relative to two 

 eclipses brought together under one cover, rendering 

 a comparison of results a matter of little labour. One 

 blemish we may, however, remark, and that is that 

 the corona reproductions are not oriented in any way. 



It may be still in the minds of our readers that, "for 



the observation of the recent eclipse of 1905, 



Admiral Chester, Commander-in-Chief U.S. Eclipse 



Squadron, was in command of four men-of-war told 



NO. 1899, VOL. 73] 



off for eclipse work in Algeria and Spain. Their 

 " station bills," showing the staff at each station and 

 the work to be accomplished, gave one a good idea 

 of the thoroughness with which the undertaking was 

 organised. We shall at any rate look forward to 

 another such volume as this, with, we hope, equally 

 successful results. William J. S. Lockyer. 



AGRICULTURE AND THE EMPIRE. 

 "jVTATURE for January 11 contains a short paper 

 ■*■ ' on a large subject. Seeing that the cultivation 

 of the soil, or agriculture, is the fundamental con- 

 dition of human existence with any approach to 

 civilisation, large is a very moderate description. 



I take it that the object of the writer was to discuss 

 the part that the Home Country should play in 

 advancing agriculture in the Empire at large. That 

 is a matter which seems to me important enough to 

 receive a little discussion. It is one with which I 

 have been a good deal occupied during the past thirty 

 years. I should like, therefore, to attempt to define 

 the present position of the problem a little more 

 precisely. 



May I begin with a very obvious remark? Agri- 

 culture is a sort of " noun of multitude." There is 

 undoubtedly only one agricultural science based on 

 physiological principles ; there are many agricultural 

 " arts " based on the application of that science, 

 whether empirical or otherwise, to widely different 

 physical conditions. The agriculture of the Lothians 

 differs widely from that of Bengal, and both differ 

 from that possible on the Gold Coast. This will 

 seem to many an absurdly trite remark. Neverthe- 

 less, experience shows that it represents a fact which 

 has often been overlooked, with loss and disappoint- 

 ment as the result. 



It may, I think, be confidently stated that arable 

 cultivation has been brought in the British Isles to a 

 pitch of perfection which is not surpassed anywhere 

 in the world. It is, however, an " intensive " and 

 highly specialised agriculture. This is readily illus- 

 trated by the yield of wheat per acre. On land of 

 prairie value, where the nitrogen removed is balanced 

 by that received from the atmosphere, it has been 

 shown at Rothamsted that the yield is roughly some 

 10 bushels or less. This actually represents the state 

 of things in the great wheat-growing countries from 

 which we draw our supplies — Argentina, Australia, 

 India and Russia — and the United States with 13 

 bushels are not much better. The yield of the United 

 Kingdom for the five years preceding 1904 was 31 

 bushels, and this was only surpassed by that of our 

 antipodal colony New Zealand, 32. 



This is largely due to the scientific research in 

 agriculture for which, I think, it may be fairly 

 claimed this country has always been preeminent. I 

 by no means think that it is exhausted. I remember 

 Sir John Lawes saying to me that, having devoted 

 half a century to the study of the soil actually culti- 

 vated, he was still absolutely ignorant as to the sub- 

 soil and the part played by it. Our knowledge of the 

 action of manures is mainly empirical, and we have 

 still to learn much of its physiological significance. 

 Without this it cannot be said that we possess a 

 rational theory of manuring. Farmers must have 

 wasted enormous sums in the application of nitro- 

 genous manures until Frankland showed that a con- 

 siderable proportion passed off unused in the drain- 

 water. 



I must confess that I am not clear that the arable 

 agriculture of the United Kingdom is in a backward 

 condition, that it does not compare favourably with 

 that of other countries, or that it stands in urgent 



