402 



NA TURE 



[August 27, 1908 



and communicated telegraphically by Prof. E. C. Picker- 

 ing ;— 



T = igo8 August 2572 



00 = 51° 9' J 



& =236° 12' \ 1908-0 



i = 145° 48' ) 



q =0103 



« =0'439S 

 Period = 2-'-55. 



Osculation, 1908 March S, igh. 45-5m. G.M.T. 



Search-ephemerides ior Comet TempeLj-Swift. — Three 

 ephemerides for the comet discovered by Tempel in 1869, 

 and recognised as periodical by Swift in 1880, are pub- 

 lished by M. E. Maubant in No. 4269 of the Asironomische 

 Nachrichten (p. 349, August 14). These ephemerides give 

 the positions of the comet from August 29 to November i, 

 the times of perihelion being taken as September 22-88, 

 Septernber 30-88, and October 8-88 respectively. The 

 following is an extract from the ephemeris for the mean 

 •date : — 



From this ephemeris it is seen that the comet is travelling 

 ■eastwards through Taurus to Auriga, and may be dis- 

 covered during the early morning before dawn. Its period 

 is about 5 J years, and it was well observed in 189 1, 

 although at its more recent returns in 1897 and 1903 it 

 was not seen. On September 9 the comet should be about 

 4° N. of /3 Tauri, w-hich rises about 10 p.m. 



Definitive Orbit of Comet 1826 V. — No. 4269 of the 

 Asironomische Nachrichten (p. 341, August 14) contains a 

 discussion, by Herr A. Hnatek, of \-ienna, of the orbit of 

 comet 1826 V, from which the autlior deduces that the 

 orbit was parabolic, the most probable ellipse giving a 

 period of nearly 28,000 years. Herr Hnatek directs atten- 

 tion to the fact that in the early hours of November 18, 

 1826, the comet grazed the sun. 



Rel.-itive Depths of the Sun-spots of a Group. — Dis- 

 cussing stereocomparator measures which he has made on 

 photographs taken at Greenwich on July 4, 5, and 6, Prof. 

 Wilhelm Krebs, in No. 4267 of the Astronomische Nach- 

 richten (p. 315, August 7), shows that the different spots 

 of the group which was then near the central meridian 

 were at different levels, and also that the changes of level 

 varied from spot to spot during the intervals between the 

 taking of the photographs. Whilst the most easterlv spot 

 showed a sharp increase of height above the datuni line, 

 the most westerly exhibited a sharp fall. The different 

 "heights, measured in 1000 km., varied from 137 to —3, 

 whilst the general increase in height during the two 

 Intervals amounted to 17,000 km., or 27 per cent. 



AN ALLEGED EXCRETION OF TOXIC 

 SUBSTANCES BY PLANT ROOTS.' 



'T'HE idea formulated a century ago by de Candolle 

 that plant roots excrete toxic substances has recently 

 ■been very much pushed forward by the American Bureau 

 of Soils to explain the effects of fertilisers and the 

 advantage of a rotation of crops. The American method 

 of experiment is to grow seedlings in water culture for 

 a few days and measure the amount of transpiration, which 

 is considered to be an index of the amount of growth. 

 The seedlings are then removed and replaced bv a second 



I (n) " Fertility of Soik a? affected by Manures." By Frank D. Gardner. 

 <L .S. Department of Agriculture, Bulletin No. 48.) 



('^) " Note on a To,\ic Suhst.ince Excreted by the Roots of Plants."-' By 

 F. Fletcher. (Memoirs of the Department of Agriculture in India, vol.ii.. 

 I^o. 3.) 



(c) "Crop Rotation and Soil Exhaustion." By F. Fletcher. (Cairo 

 ocientific Journal, vol. ii.. No. 19.) 



KG. 2026, VOL. 7S] 



batch, without changing the water ; the rate of transpira- 

 tion is found to be diminished, showing (it is stated) 

 that a toxic body excreted by the roots of the first batch 

 is adversely affecting the second. Further, seedlings grown 

 in an aqueous extract of certain poor soils are found to 

 transpire less water than others grown in distilled water, 

 and it is concluded that these soils contain some toxic 

 material, presumably excreted by plants. The toxic body 

 is, however, precipitated on addition of charcoal, ferric 

 hydrate, and solutions of various manures ; and the Bureau 

 of Soils argues that the function of fertilisers, in some 

 cases at any rate, is not to feed the plant, but to precipitate 

 the toxin e.xcreted by previous plants. Rotations of crops 

 are of advantage, because the toxin excreted by one plant 

 is not necessarily harmful to plants of a different order. 



It cannot be said that any very convincing evidence is 

 offered in support of this view. The assumption that 

 ti-anspiration is a measure of plant growth is not borne 

 out by any of the figures quoted ; thus in a series of 

 experiments given in Bulletin No. 3(5 the crop weights and 

 transpiration results are : — 



Another weak point is that the experiments are made 

 with seedlings, and last only a few days, instead of being 

 carried on to the end of the plant's life. The nutrition of 

 the seedling is not the same as that of the plant, and 

 even if it wei-e demonstrated that secretion from seedling 

 roots took place, it would not follow that there was a 

 similar secretion from the roots of fully grown plants. 



In the last Bulletin from the Soil Bureau (No. 48) an 

 account is given of more than 13,000 pot trials with soils 

 fi-om different parts of the United States. The results 

 show, as might be expected, that addition of manures in- 

 creases the crop, and that each manurial substance exerts 

 a specific effect which is not shown by any other ; with 

 this statement everyone would agree. The further con- 

 clusion is di-awn that the character of fertiliser required 

 depends more upon local conditions and practices than on 

 the type of soil or the geological formation to which it 

 belongs, so that the fertilisers required for the same type 

 of soil as It occurs in different localities usually vary more 

 than those required for very different types when in the 

 same locality and subjected to similar environment. If 

 this generalisation turned out to be true, it would be more 

 easy to reconcile with the plant excretion view than with 

 the nutrition view of the function of fertilisers, but an 

 examination of the tables does not show that there is any 

 proof. Averages are taken without any regard to their 

 probable value. Thus in one section of the table we find 

 three soils only, and they give the following percentage 

 increases when ti-eated with various manures, yet the 

 author finds no difficulty in taking an average : — 



The magnitude of the experimental error can only be 

 inferred from one table, where the separate crop weights 

 for twenty pots are given ; it would appear to be consider- 

 able, since the weights vary from 58-7 grams to 89-9 

 grams ; but the author groups the pots in sets of five, and 

 in this way reduces the error to 5 per cent., which is given 

 as the probable error for all the experiments ! There 



