380 



SCIENCE 



[N. S., Vol. XL., No. 1028 



The question of whether the azotobaoter both 

 fix the atmospheric nitrogen and convert it into 

 nitric acid, respectively nitrates, or whether this 

 latter work is done wholly by another genus or 

 other genera of bacteria is, perhaps, a question to 

 be settled, but, be it settled as it may ... we 

 have instances of the accumulation of very large 

 quantities of nitrates in our soils always associ- 

 ated with the brown color which we know to be 

 caused by the azotobacter. I believe, and this be- 

 lief is based upon tentative facts, that the azoto- 

 bacter are at the same time nitrifiers, i. e., that 

 they possess a double function, which, I believe, 

 has already been asserted, but not generally ac- 

 cepted. 



This subject has not been referred to in later 

 bulletins because I believe that my tentative 

 facts veere interpreted vsrongly. 



The tentative facts referred to were, in the 

 first place, that vyith pure cultures made on 

 sand I obtained a decided color reaction with 

 phenolsulfonic acid which might readily be 

 taken for the reaction due to nitric acid, in 

 the second place, the power of pigmentation 

 in successive cultures of azotobacter weakens 

 and finally disappears. The loss of this power 

 of pigmentation is not permanent, for, as Pro- 

 fessor Sackett has since shown, the addition 

 of a very small amount of a nitrate to the cul- 

 ture medium restores it. 



I interpreted the former fact, the reaction 

 with phenolsulfonic acid, as rather strong 

 proof of the presence of nitric acid and the 

 latter fact as supporting this view. It seemed 

 to me that the second fact given, i. e., the 

 weakening of the power of pigmentation, 

 pointed to an ability of the azotobacter to 

 nitrify in a limited measure and that this 

 function was lessened in the succeeding gen- 

 erations grown on mannite-agar until it finally 

 vanished while the purely vegetative function 

 was retained apparently unimpaired. With 

 these facts and views in mind I wrote the sen- 

 tences quoted from Bull. 178 of this station but 

 I was very far from being satisfied with the 

 tentative facts. At my request Professor Sack- 

 ett kindly made other cultures from two of his 

 stock cultures which had shovsm marked abil- 

 ity to form pigments. These cultures were 



made on a much larger scale than those previ- 

 ously made on sand and were allowed to incu- 

 bate till the pigments were well developed. 

 The membranes were removed from the agar 

 and the agar washed with distilled water. The 

 wash water was rendered alkaline by the addi- 

 tion of sodic carbonate and evaporated to dry- 

 ness. The membrane that had been removed 

 was added to the residue and the whole was 

 thoroughly mixed and dried. A portion of the 

 dried mass was tested with phenolsulfonic acid 

 and yielded a deep brovni solution which, on 

 sufiicient dilution, gave a yellow color with a 

 tinge of brown. A most excellent imitation 

 of those unsatisfactory solutions sometimes ob- 

 tained on applying this test to samples of soils. 

 We tried such means as were at our command 

 to remove the brown color or tinge which does 

 not belong to the nitric acid reaction but with- 

 out success. We rejected this phenolsulfonic 

 acid test because the results were so doubtful 

 that we considered them valueless in this par- 

 ticular case. 



A larger portion, in fact all that we had left 

 of the dried membrane, was treated with 

 ferrous ehlorid and hydrochloric acid with all 

 of the precautions demanded by this method. 

 The volume of gas evolved was only 2.3 c.c. 

 which was transferred to an absorption burette 

 and a freshly boiled, concentrated ferrous 

 ehlorid solution allowed to flow into the gas. 

 No absorption took place and no brovsra color 

 was produced on the margins of the slowly in- 

 flowing stream of ferrous ehlorid solution 

 which constitutes an exceedingly delicate test 

 for nitric acid. These residts indicated that our 

 previous caution was fully justified and that 

 the color obtained with phenolsulfonic acid 

 was due, not to nitrates but to the action of 

 the reagents upon the substances in the mem- 

 brane itself, most probably upon the pigments. 



We may add apropos to these pigments that 

 while they are difficultly soluble or insoluble in 

 the menstrua usually used, pure water, alcohol, 

 etc., the presence of various salts in aqueous 

 solution cause them to dissolve to a greater or 

 less extent; one, which, in some cases, is suffi- 

 cient to impart a yellowish-brown color to the 

 solution. We have often met with this in ma- 



