330 



NA TURE 



[February 6, 1908 



The rude pictographs or petroglyphs found in the island 

 are equally remarkable. Some of the best specimens are 

 engraved on boulders in rivers or in their vicinity, thus 

 possibly indicating their use in some mimetic rain cere- 

 mony. Others, drawn with less care, differing from the 

 river sculptures in size, shape, and apparently in signifi- 

 cance, are inscribed in the caves so common in the 

 calcareous rocks of the island. They represent in some 

 cases a head furnished with horns, human faces, and rude 

 representations analogous in type to the " three-pointed " 

 images, with circles, spirals, and other symbols. Our 

 knowledge of the aboriginal cults is at present too scanty 

 for any attempt to explain their significance. That they 

 represent objects of religious worship seems fairly certain. 



Porto Rico is thus a most promising field for archaeo- 

 log-ical investigation. The report of Dr. Fewkes suggests 

 many interesting problems. It is well written, and, like 

 other publications in the same series, is admirably illus- 

 trated. 



Another article by the same writer describes a tour of 

 exploration in eastern Mexico, undertaken with the object 

 of ascertaining the relationship between the mound builders 

 of the lower Mississippi and the people of the Mexican 

 coast known as Totonac and Huaxtec. The Totonac 

 metropolis, Cempoalan, was for the first time examined. 

 From this investigation it seems probable that the mound 

 builders of the southern States were more closelv con- 

 nected with the races of eastern Mexico than with those 

 of the arid region of the south-west or even with the 

 plateau tribes of eastern Mexico. 



The third part of the thirteenth volume of the Journal 

 •of the Academy of Natural Sciences, Philadelphia, is de- 

 voted to a fresh exploration of Moundville, Alabama, and 

 sites on the Crystal, Chattahoochee, and Lower Flint 

 Rivers, and the Ten Thousand Islands of Florida, by Mr. 

 C. B. Moore. The appearance of the Swastika symbol 

 .and other objects discovered at Moundville seems to indi- 

 cate that this was an important religious centre, specially 

 devoted to the worship of the sun, conducted in temples 

 by an order of priests, who kept the sacred fire continuallv 

 burning as an emblem of the luminary. The discovery at 

 the Crystal River of an ear-ornament overlaid or covered 

 with meteoric iron is an interesting proof of the skill of 

 this race in metallurgy, and it seems to connect the races 

 of Florida, who buried their dead in mounds, with the 

 people of the Ohio valley region. This report also is 

 provided with excellent illustrations of the remarkable 

 series of finds which were secured by Mr. Moore's 

 expedition. 



THE NITROGEN PROBLEM IN AGRICUL- 

 TURE.' 

 "pOR many years what is known in agriculture as the 

 nitrogen problem has received considerable attention 

 both from men of science and from practical men. It 

 Tias two aspects. Few soils contain nitrogen compounds 

 In suflicient quantity for the needs of non-leguminous 

 crops, and the application of nitrogenous manures is one 

 of the commonest, as well as one of the more costly, 

 operations of modern agriculture. On the other hand, 

 leguminous crops not only need no nitrogenous manure 

 for themselves, but actually increase the store of nitrogen 

 compounds in the soil, and dispense with the necessity of 

 adding more for the succeeding crop. The problem would 

 obviously vanish if leguminous crops could be grown every 

 ■other year, but unfortunately they are liable to " sick- 

 ness," and can only be grown once in four or even six 

 years. Even as it is, however, any method that increases 

 the nitrogen-fixing power of a leguminous crop is a 

 welcome addition to the resources of a farmer. 



In t886 Hellriegel and Wilfarth showed that nitrogen- 

 fixation is the work of certain micro-organisms associated 

 with the leguminosae. It has since been shown that 

 they can grow apart from the plant, and can be inoculated 

 Into soils, and also that an increased crop may follow 

 such inoculation. 



On three occasions cultures of these organisms have 

 '^ " Si-ed and Snil Inoculation For leguminous Crops." By Prof. W. B. 

 Bottomley. (London: Coimtiy Life Office.) 



NO. IQ97, VOL. ']']'\ 



been widely distributed among farmers. In 1896 Nobbe 

 sent out " nitragin " ; in 1903 Moore's cultures were 

 issued in America; and now we have Prof. Bottomley 's 

 cultures. Each time very great and widespread interest 

 has been aroused, the matter has been discussed at length 

 in the daily Press, and has even on occasion formed the 

 subject of questions in the House of Commons. The 

 subject appeals to almost everyone. Few scientific 

 problems are more interesting than the wholly unparalleled 

 synthesis of complex organic compounds from free nitrogen 

 and other simple bodies effected by these micro-organisms 

 at the low temperatures of the soil. The practical man 

 sees in inoculation the possibility of increased leguminous 

 crops and of less expenditure on nitrogenous manure for 

 his other crops. The man in the street, who has always 

 been ready to take an interest in nitrogen since S'ir 

 William Crookes's British Association address at Bristol 

 in 1S98, sees the threatened nitrogen famine averted and 

 his food supply rendered secure for a long time to come. 



Inoculation has proved very successful on virgin soils, 

 or in dealing with new leguminous crops, but there is 

 little or no evidence that it is effective when the soil \i 

 already in cultivation and the crop no longer new. 

 Nobbe's cultures failed, and Moore's cultures were not 

 particularly successful when applied in ordinary farm 

 practice. Certain other less boomed cultures, e.g. 

 Hiltner's, have done better, and have sometimes given 

 20 per cent, or 30 per cent, increases in crop. Prof. 

 Bottomley tested his culture by distributing more than a 

 thousand specimens ; 80 per cent, of the reports received 

 (unfortunately we are not told the actual number) showed 

 an increase in crop. The pamphlet before us consists 

 largely of extracts from these reports. 



Experiments of this nature are very difficult to carry 

 out. Great care is necessary in selecting the ground, 

 proper control plots are needed, and the experiment must 

 be continued for several years without essential modifica- 

 tion. Under favourable conditions, the error of a field 

 experiment lasting many years may be as low as 5 per 

 cent., but for shorter periods it is much higher. We 

 cannot find any evidence that these facts have been taken 

 into account ; on the contrary, most of the trials have 

 obviously been made by novices. Only in a few cases 

 have any weights been taken, and the results have usually 

 been guessed. Here is an " experiment" with peas: — 



'* I planted the inoculated peas on land that had not 

 been manured for manv years, and had a crop of peas 

 quite equal to those grown by a friend on manured soil " 

 (p. 24). 



Another : — " ' Gradus ' without inoculation, a fair crop, 

 but they were soon over." 



" ' Sutton's A I ' inoculated, heavy crop, with abundance 

 of well-filled pods " (p. 21). 



Again : — " First sown peas, inoculated, a fine crop. 

 . . . Second sowing, uninoculated, results very poor. . . . 

 The ground on which first crop was sown had had no 

 peas on it for several years, whereas the ground on which 

 second crop was sown had had peas grown on it in the 

 previous year " (p. 17) 



It is scarcely necesary to say that statements of this kind 

 are of no scientific value. 



Fifty-two reports are quoted ; fifteen must certainly be 

 rejected, eighteen give no figures, thirteen give estimated 

 figures, and six give weights. Confining our attention 

 to the nineteen cases \\*here figures are given, we find 

 that in four the increased yield is 20 per cent, or less ; 

 this cannot be much greater than the magnitude of .the 

 experimental error. In nine cases the estimated increase 

 is 30 per cent, to 50 per cent., and in five cases more than 

 50 per cent. A detailed discussion is obviously impossible ; 

 we can only point out that a few experiments properly 

 made on sound lines would have been very much better 

 than all these testimonials. So far as they prove anything 

 at all. however, they go to show that the author has, 

 like Hiltner in Germany, Moore in the United .States, and 

 Golding ' here, obtained cultures which may in suitable 

 cases increase the vield of leguminous crops, and the 

 increase seems to be estimated in the majority of cases 

 at about 30 per cent, to 50 per cent. 



Now if it this turns out to be correct it will be a very 

 useful result, but in ordinary farming it can only be 



