282 



SCIENCE 



[Vol. LVI, No. 1445 



(6) When the reaction of acidified toxin is 

 brought back to neutral (before dilution), the 

 resulting solution has the titer of the original 

 toxin. The I'e-aeidification and re-neutraliza- 

 tion can be accomplished several times in suc- 

 cession with the result that acidification inva- 

 riably increases the potency and neutralization 

 returns it to the original titer. 



(7) When one cubic centimeter of the 

 original toxin has been distributed through a 

 sufficiently large amount of a diluent so that 

 not every cubic centimeter of the diluent would 

 be theoretically expected to contain one 

 molecule of original toxin, the results of the 

 injection of such highly diluted toxin become 

 irregular and apparently depend on the pres- 

 ence or a'bsence of a molecule or a small num- 

 ber of molecules of toxin in each portion 

 injected. 



(3) While not every filtrate yielded an 

 equally potent product on acidification, the 

 same filtrate consistently titrated as indicated, 

 even though repeated tests were performed 

 several days and weeks apart. 



On the other hand, it is evident that there 

 are a number of considerations militating 

 against the validity of our observations. 



(1) Using the same strain of the organism 

 and similar culture medium, it was not always 

 possible to obtain the same degree of increase 

 in potency of the acidified filtrates, although 

 in all cases some increase was observed. Ap- 

 parently, the uncontrollable differences in com- 

 position of the culture medium during the 

 early growth of the organism has something to 

 do with the degree of change in potency which 

 the toxin will undergo upon its subsequent 

 acidification. 



(2) It is difficult to conceive how such a 

 small number of molecules which can theoret- 

 ically be expected to be present in the small 

 amount of toxin injected can produce the 

 effect. 



(3) The extreme simplicity and low molecu- 

 lar weight which the active substance seems to 

 possess according to calculation is difficult to 

 reconcile with its strict biologic specificity 

 which would postulate a more complex struc- 

 ture. 



(4) We find that while this active toxin is 

 neutralized by a specific antitoxin, the neu- 



tralization does not go according to the law 

 of multiple proportions, but is in fact more 

 efficient. 



(5) It is difficult to explain why such a 

 simple molecule as that which the active toxin 

 seems to possess can not pass bacterial filters 

 which are comparatively permeable to the 

 original toxin. 



These, as well as other considerations, indi- 

 cate that a further study of the subject is 

 necessary. We feel more inclined to believe, 

 and some of our most recent observations 

 strengthen this belief, that while the toxin does 

 unquestionably undergo an increase of potency 

 under certain conditions of the experiment, the 

 degree of this increase probably is not as great 

 as some of our findings seem to indicate. We 

 suspect that there may occur an ultramiscropic 

 precipitation of the toxin-carrying portion of 

 the medium. If the minute particles of such 

 a precipitate should possess particularly high 

 adsorptive power, they could be carried from 

 dilution to dilution and thus vitiate the accu- 

 racy of the calculation. Since circumstances 

 force us to interrupt this study for the time 

 being, we thought it worth while to call the 

 attention of other workers to this interesting 

 phenomenon. With this in view, we are pre- 

 paring detailed protocols of the experiments 

 to date which we hope to publish in the near 

 future. 



J. Bronpenbeenner 



Harvard Medical School, 

 Boston 



SOIL SHIFTING IN THE CONNECTICUT 

 VALLEY 



Eecentlt two articles have appeared in the 

 current volume of Science (Nos. 1413 and 

 1426), reporting soil shifting by wind. In the 

 Connecticut Valley in the vicinity of Amherst 

 sand storms are a common occurrence. The 

 prevailing winds, coming from the northwest, 

 have a rather uninterrupted sweep down the 

 valley, and at times pick up and transport 

 large quantities of soil consisting of sands, 

 sandy loams, silt loams, clay loams and clays. 

 It is the coarser members of the above soils 

 that are most eroded because the heavier soils 

 are usually covered with vegetation. 



This shifting of the soil by wind action has 



