2l8 



NA TURE 



[December 23, 1909 



tion Prof. Dendy puts forward as to its function is quite 

 a reasonably probable one. 



Geo. E. Nicholls. 

 King's College, December 17. 



Nitrogen-fixing Bacteria and Non-Leguminous Plants. 



I REGRET that owing to Mr. A. D. Hall's letter of 

 July 12 (Nature, July 22) appearing during vacation, I 

 had not an opportunity of seeing it until my return to 

 college at the commencement of term. I trust that, in 

 spite of the long delay, I may yet be allowed to reply to 

 some of Mr. Hall's questions. 



Mr. Hall commences by stating that my conclusions are 

 based " on experiments to show that Pseudomonas, the 

 bacterium associated with the leguminous plants, will fix 

 more nitrogen, &c." This is not correct, for it is definitely 

 stated in the abstract that both the Pseudomonas and the 

 Azotobacter used for the nitrogen determination results 

 quoted were obtained from a non-leguminous plant — the 

 root tubercles of Cycas. 



There are four non-leguminous plants possessing root 

 tubercles which contain nitrogen-fixing organisms. In all 

 four a species or variety of Pseudomonas is present, but 

 in Cycas only is Pseudomonas found living outside the 

 cortical cells, in the algal zone, and in Cycas only is 

 Pseudomonas found in association with Azotobacter. As 

 stated in my paper, the nitrogen determinations quoted 

 had reference only to these organisms from Cycas, and 

 were made in order to determine to what extent, if any, 

 Cycas Azotobacter assisted Cycas Pseudomonas in assimil- 

 ating free nitrogen. 



" I^seudomonas, the bacterium associated with the 

 leguminous plants," and the use of " a reasonably active 

 culture " of Azotobacter, to quote Mr. Hall, had nothing 

 to do with the determination experiments. On what 

 grounds, then, does Mr. Hall state that " the only con- 

 clusion that could be drawn from Prof. Bottomley's figures 

 would be that Pseudomonas injuriously affects the power 

 of .Azotobacter to fix nitrogen "? Naturally, a determina- 

 tion of the amount of nitrogen fixed by the Azotobacter 

 alone was made Cfor Mr. Hall's information I may say 

 that it was very small, —0-56 mgr.), but as the immediate 

 object of the experiment was to determine the effect of 

 Cycas Azotobacter on Cycas Pseudomonas, it was not 

 thought necessary to quote the figures for Cycas Azoto- 

 bacter alone, especially as this will be dealt with in a 

 future paper. 



Again, when spealcing of the experiments with oats, Mr. 

 Hall says that the mean error of + 10 per cent, found in 

 Rothamsted experiments " would more than cover the 

 differences observed by Prof. Bottomley's experiment with 

 oats." The figures given for oats are: — average weight 

 per plant, untreated, 0-42 grm. ; treated, 0-74 grm. ; 

 increase, 0-32 grm., or 76 per cent. Surely Mr. Hall 

 cannot have read the abstract carefully or he would not 

 have stated that a mean error of + 10 per cent, more than 

 covers an increase of 76 per cent. ! As regards probable 

 experimental error, the abstract states that the oats were 

 " grown in sand, dressed with phosphates, potash and 

 lime." Why, then, should Mr. Hall assume that they 

 were grown " in soil which presumably already contains 

 both organisms "? 



As regards the field experiment with barley, duplicate 

 samples were not taken from different parts of the un- 

 treated plot, but the total yields from treated and untreated 

 plots were kept separate, and the treated showed an 

 increase of 13-6 per cent. The sample for estimation of 

 nitrogen content was taken from the bulk in each 

 case. 



The treated bulbs gave an increased yield of 18-6 per 

 cent. They were not selected as being specially " suitable 

 for experiments on nutrition," but to determine the effect 

 of the mixed culture of bacteria on a totally different kind 

 of plant from any of the other experiments. The bulbs 

 were already planted when the experiment commenced, 

 hence the original weight of the bulbs is not available. 

 As, however, the bed contained 500 bulbs, and was divided 



NO. 2095, VOL. 82] 



into two equal halves with 250 bulbs in each, one may 

 assume that the original weight in each half was approxi- 

 mately equal. 



As regards the culture solution " being a considerable 

 factor in any beneficial effect experienced," it is difficult 

 to comprehend how 45 grs. potassium phosphate and J gr. 

 magnesium sulphate dissolved in a gallon of water for 

 the culture solution, and this afterwards diluted with fifty 

 gallons of water before applying, could possibly produce 

 any appreciable effect on growing crops. Assuming that 

 the bacteria had not used up any of the culture salts in 

 their growth to produce the culture solution, can one 

 imagine any benefit to plants by watering them with a 

 solution containing 1/50,000 part of potassium phosphate 

 and 1/450,000 part magnesium sulphate? 



Mr. Hall appears to consider his questions " somewhat 

 critical." True and fair criticism is always welcome in 

 scientific investigation, for it is only thus that one can get 

 at the truth of things, but misstatements of fact or mis- 

 representations of results exceed the bounds of criticism. 

 So far as Mr. Hall correctly states the experiments and 

 results given, one welcomes his criticisms. 



W. B. BOTTOMLEY. 



King's College, Strand, W.C, 

 November 25. 



In Prof. Bottomley's paper in the Proceedings of the 

 Royal Society I cannot read that he makes any claim that 

 the Pseudomonas and the Azotobacter he isolated from 

 Cycas were in any way different in kind from the usual 

 forms of these organisms. When I used the expression 

 " Pseudomonas, the bacterium associated with leguminous 

 plants," it was as a sort of explanatory label for the lay 

 reader, but if it has confused the issue at all I would 

 still repeat my former criticism on Prof. Bottomley's basis 

 that the Pseudomonas and Azotobacter he used are special 

 and unlike all others. The question was if " the associa- 

 tion gave an increased power of assimilating free 

 nitrogen," and Prof. Bottomley's figures are: — 



Conirol... 



Pseudomonas alone ... 

 Pseudomonas f Azotolacter , 



0-48 nigm. nitrogen 



0-91 



I '24 



and I asked if the action of .Azotobacter alone ought not 

 also to be known before any answer is possible. Prof. 

 Bottomley now tells us that Azotobacter alone fixes 

 056 mgm., so that I still conclude that Pseudomonas and 

 Azotobacter together {1-24) are less effective than when 

 grown separately (o-gi -f 0-56), were I not more inclined 

 to think that all the differences are within the limits of 

 experimental error. 



To pass to the experiments with plants, by an error 

 which the context rendered sufficiently obvious I wrote 

 " oats " instead of barley when dealing with Prof. 

 Bottomley's first-quoted experiment with soil. The oat 

 experiment is beside the point ; it only demonstrates fi.xation 

 of nitrogen by .Azotobacter and Pseudomonas — a fact on 

 which we are all agreed ; but when Prof. Bottomley claims 

 increased crop production due to inoculation with the two 

 organisms in the open ground, we do want the means of 

 judging what weight to attach to the results. He gives 

 for the first barley experiment a gain of 13-6 per cent. ; in 

 barley 2 the increase cannot be estimated ; the Galtonia 

 experiment shows i8-6 per cent, increase, the parsnips 

 21-7 per cent. Such percentage differences are not outside 

 the limits of error for a single plot experiment worked 

 on a large scale with every condition in favour of accuracy ; 

 when they are differences between lots of 250 bulbs or 68 

 parsnips I have no hesitation in regarding them as with- 

 out any significance whatever. I have just had pulled 

 and weighed two lots of fifty successive roots of man- 

 golds growing side by side in neighbouring rows on the 

 same plot, inside rows perfectly similar to the eye and on 

 a very uniform plot, yet one lot weighed 220 lb., the other 

 lot 176 lb., a difference of 25 per cent. The assistant who 

 gathered the roots was quite unaware of the question in- 

 volved ; his instructions were to begin inside the plot and 

 take fifty consecutive roots along a row, then another fifty 



