B y F . J . Al w a y a n d R . S . Tr u m b u l l.
T h e developm ent of so-called “ dry fa rm in g” or
“ dry-land fa rm in g” on the semi-arid portions of the prairies of N orth A m erica has introduced problem s in soil fertility distinct from those of humid or of irrigated soils. The characteristic feature of d ry
land farm ing, in so far as it relates to the soil, is the practice of bare fallow ing to store w ater in the sub
soil, a practice,* well know n to cause, in humid regions, a rapid decline in soil nitrogen, the portion removed in the form of crops being only a small p art of the whole that is lost. W hile in the least arid portions of the semi-arid regions certain intertilled crops m ay be found capable of taking the place of the fallow to some extent, such a change in cultural m ethods is not likely to lessen in an y considerable degree the rate of loss of nitrogen. T h e use of legum inous crops as green manures has been advised as a substitute for the fallow .1 T he growth of the legumes, however, e x hausts more or less of the moisture which fallow ing would conserve, thus increasing the supply of nitrogen a t the expense of that of w ater and so defeating the ve ry object of the fallow.
Investigations in dry-land farm ing on the G reat Plains have been undertaken only so recently b y e x perim ent stations in the U nited S tates th at d ata are not y e t available to indicate how serious the nitrogen problem is, how soon it will becom e acute, or how best to solve it. For tw en ty-tw o years, however, an e x perim ental farm has been maintained b y the Canadian governm ent a t Indian H ead, Saskatchew an, and during the whole of this tim e the practice of bare fallow ing has been em ployed for the conservation of moisture, it having been introduced into the Canadian N orthw est b y Mr. Angus M ackay, who has been
1 C hilcott, “ D ry -lan d F a n n in g in th e G re a t P lains A re a .” Y ear-b o o k , U . S. D e p ’t of A gr., 1907.
T H E J O U R N A L O F I N D U S T R I A L A N D E N G I N E E R I N G C H E M I S T R Y . April, 1910 others. Considering the five w idely separated points o f Indian H ead in Saskatchew an, N orth P la tte in reports of the Canadian Experim ental Farm s and p artly from private inform ation furnished b y Mr. A. M ackay.
A L W A Y A N D T R U M B U L L O N N I T R O G E N P R O B L E M U N D E R D R Y F A R M I N G . th ey have appeared to have reached their heaviest
growth.
T h e m ost striking result of the analyses is the lack of any apparent connection between the composition of the samples and the treatm ent received b y the different plots during the last nine years. No. i, the sample highest in humus, nitrogen and carbon, is from a plot on which three crops of legumes have been plowed under, while the samples showing the least of the same constituents, viz., 8 and 9, are likewise from plots on which three crops of legumes had been plowed under. A com parison of the averages shows no distin ct difference between the fifteen plots on which legum es had been used as green manure and the other seven.
T h a t an explanation of the differences between the different plots is possible is evident from the data in T able III . Suspecting that distin ct differences m ight be found between different individual samples from the sam e plot, according as to whether they were from the till or from the lacustral clay, ten individual samples from rotation plot 14 were kep t separate and analyzed.
T ill forms the surface of the w est third of this plot and of an isolated area near the center, while the east end has a surface of typical lacustral clay. Near the edge of this area of lacustral clay and about tw en ty feet w ithin it, five samples, 24 to 28, were taken at intervals of ten feet in a north and south line, and, be
ginning tw en ty feet w est of 26, five samples, 29 to 33, were taken a t intervals of eight feet in an east and west line. W hile 29 and 30 are really typical of neither type, being interm ediate between the two, 31 to 33 are quite representative of the till, being from the isolated area above mentioned, and 24 to 28 are fair samples of the lacustral clay. ,
Ta b l e I I I . — Hu m u s, Ni t r o g e n a n d Or g a n i c Ca r b o n o f In d i v i d u a l Sa m p l e s f r o m t h e Sa m e Pl o t.
H u m u s N itrogen.
O rganic carb o n .
Soil N o. P e r cent. P e r cen t. P e r cent.
24 L a c u stra l clay ... 0.291 3 .2 7 25 “ “ ... . 2 .2 6 0 .2 6 8 3 .1 9
26 44 “ ..., . 2 .2 9 0 .2 8 2 3 .0 1
27 “ “ .... . 2 .2 8 0 .2 7 8 3 .1 8 28 44 " ... . . 2 .2 8 0 .2 7 8 3 .0 8 A verage la c u stra l c la y ... 0 .2 7 9 3 .1 5 29 B ou ld er c la y ... . . 2 .0 3 0 .2 7 8 2 .8 5
30 , . 1.9 4 0.2 6 3 2 .7 7
31 “ “ ... . 0.231 2 .4 3
32 “ “ ... 0 .2 2 7 2 .2 9
33 “ * ' ... .. . 1.57 0.221 2 .3 3 A verage B o u ld e r c la y ... . 1.7 7 0 .2 4 4 2 .5 3
T he samples of lacustral clay are richer in hum u total nitrogen and carbon than are those of till. From the differences found betw een the two types of soil on the same plot, it seems probable th at the differences found betw een the different plots should be attributed chiefly to the varyin g am ounts of the two types of surface soil.
Sam ples were secured from other fields and also from tree rows which had for the preceding 15 years been under continuous bare cultivation. These rows
of trees were planted along the lanes in 1892 and since that time the soil for a distance of tw elve feet on each side of the trees has e very summer been plowed and cultivated sufficiently to prevent the growth of grass or weeds. A sum m ary of the results are given in T able IV .
Ta b l e I V . — Re l a t i o n o f Co m p o s i t i o n o f Su r f a c e Si x In c h e s o f So i l t o Cu l t i v a t i o n.
T o ta l O rganic
No. of H u m u s. N itrogen. carbon.
sam ples. C u ltiv a tio n . P e r c en t. P e r cen t. P e r c en t.
6 V irgin p rairie... 2 .8 4 0 .3 8 4 4 .2 0 2 In c u ltiv a tio n 7 y rs., in grass 18 y r s ... 2 .7 5 0 .3 8 4 4 .0 5 2 “ M 16 11 “ " 9 y rs ... 2 .4 6 0 .3 6 7 3 .8 2 2 " 21 ** “ " 4 y rs , , . 2 .2 5 0 .3 0 3 2.10
22 •* 44 25 44 (R o ta tio n plotfield) 2.01 0 .2 6 6 2.68
6 Tree strip s; in o rd in a ry cu ltiv a tio n
10 y ears, in b a re c u ltiv a tio n 15 y rs . 2 .12 0 .2 7 4 2 .8 4 6 In cu ltiv a tio n in fields a d ja c e n t to tree
s tr ip s ... 2 .3 4 0 .3 2 6 3 .3 5
I t is impossible from the data on hand to satisfac
torily estim ate the original composition of the virgin prairie where lacustral clay occurs.1 The numbers given in the first line of the table are the averages of samples from the boulder clay and from the lacustral clay summits. The numbers are certainly too low because samples from some fields long under cu ltiv a tion give slightly higher percentages than these.
On soil where grass (in this case m ixed with clover) has grown for 18 years, follow ing 7 years of cu ltiv a tion, no distinct loss is to be observed. T h e longer the fields have been in grass the less has been the decline. W here no grass a t all has been grown the loss is greatest. Continuous bare cu ltivation in the tree rows has caused greater losses than where bare cultivation (fallow) and cropping have alternated.
T h e extrem e loss is probably about one-third of the original am ount of all three constituents.
T h e data showing the effect upon the yield of w heat produced b y replacing the sum m er-fallow b y a legu
minous crop plowed under as a green manure have been obtained chiefly from the field of rotation plots described above (Table II). A sum m ary of the results is given in T able V . Judged b y its crop-pro
ducing power, this field was very uniform in 1899, the yield of the 12 plots of w heat ranging only from 35 bushel 20 lb. to 36 bushel 16 lb. per acre. T h e data for w heat after legum es in 1899 are from a series of half
acre plots in another field. On account of the little d ata on the yields after fallow in the rotation field there are given in the table the averages of all field lots of the standard varieties grown on summer- fallowed land. There being no field or half-acre plots of w heat after fallow in 1900 it has been necessary to em ploy the yield of a sm aller plot. In 1902 the w heat after fallow on the three rotation plots was in
jured so seriously b y rust th at the yield was lower (20.3 bushel per acre) and the qu ality poorer than on any other plot or field on the farm. Accordingly this yield should not be used in such a comparison.
1 A lw ay a n d V ail, Th i s Jo u r n a l, 1 , 17 (1909).
138
Ta b l e V . — Yi e l do f Wh e a t p e r Ac r ea t In d i a n He a d a f t e r Fa l l o w, a f t e r Le g u m e s Pl o w e d u n d e r, a n d a f t e r Wh e a t.
A fte r
A fte r fallo w . le g u m in o u s A fte r i ■--- * g re e n w h e a t, o a ts
C h ec k p lo ts . m a n u re s . a n d b a rle y . F ie ld lo ts. ,--- s «--- ■--- . ■
-Y ear. A cres. Y ield , bu.
N o. oi 1 a c re Y ield,
p lo ts. b u . N o o f
£ a c r e Y ield, p lo ts , b u .
N o. o f
£ a cre p lo ts. Y ield ,
bu.
1S99 16 3 2 .7 13 3 5 .9 6 3 4 .5 .1 2 8 .9
1900 0 12 2 7 .0 5 1 4 .9 4 5 .4
1901 17 4 9 .2 1 4 4 .7 5 43 .1 2 3 8 .4
1902 40 3 3 .4 3 __3 5 2 4 .9 4 2 5 .5
1903 26 3 5 .8 0 5 2 3 .8 4 15 .9
1904 44 3 9 .3 1 3 6 .0 5 3 4 .5 2 3 0 .2
1905 44 4 0 .0 3 3 5 .3 5 3 0 .0 4 23 .1
1906 45 3 0 .5 0 5 2 0 .7 4 16.5
A verage, 3 4 .6 2 8 .3
,
2 2 .9In two years, 1899 an^ 1901, when the previous harvest had been followed b y excep tionally w et weather, the yield after legum es was about the same as that after fallow , the autum n rainfall probably having been sufficient to fu lly charge w ith moisture the subsoil of all the fields. In the other years the legumes, b y exhausting the w ater of the subsoil, lessened the yield of the succeeding crop. W heat has given a lower yield when sown on w heat stubble than after legumes, chiefly because the soil has been left drier in the autum n, although p artially, probably because the supply of available nitrogen has been less.
E ven in the case of the two w et years mentioned, the difference in yield is considerable.
The d ata in T able V I seems sufficient evidence that, notw ithstanding a decline of 30 per cent, or more in soil nitrogen, this elem ent has not y e t become a lim iting factor in the yield of w heat on fallowed land at Indian H ead. T h e table gives the yield of R ed F ife w heat both w ithout an y fertilizers and w ith 100 lbs.
and 200 lbs., respectively, of sodium nitrate. The plots were on fallowed land in every case, soil not previously fertilized being taken each year. In 1903 all the plots in the fertilizer experim ents were so b adly injured b y rust th at the crop w as not weighed. The average yields of the three plots for the rem aining six years were p ractically the same. No trials are reported of the use of nitrates on w heat sown on stubble.
Ta b l e V I.—Yi e l d s o k Wh e a t a t In d i a n He a d w i t h a n d w i t h o u t t r o g e n Fe r t i l i z e r s.
N o fertilizer. 100 lbs. N a N 0 3. 200 lbs. N a N 0 3.
S traw , G rain , S traw , G rain , S traw . G rain ,
Y ear. cw t. b u . cw t. bu . cw t. bu .
1900 1 7.4 2 7 .7 17 .6 2 7 .3 17 .6 2 7 .3
1901 66.0 6 2 .7 5 9 .2 6 1 .3 5 4 .8 5 8 .7
1902 3 7 .6 2 9 .2 3 1 .6 2 8 .0 3 8 .8 3 0 .7
1904 2 8 .6 2 6 .3 3 0 .4 2 8 .7 2 6 .4 2 3 .0
1905 7 2 .4 4 2 .7 6 2 .8 4 2 .0 5 4 .0 4 3 .3
1906 4 8 .0 3 5 .3 6 2 .4 3 8 .0 5 2 .4 3 8 .0
A verage, 4 5 .0 3 7 ,3 4 4 .0 3 7 .5 4 0 .7 3 6 .9
There are two distinct causes of a lowering of the content of organic m atter, humus, and nitrogen of cu ltivated soils a t Indian Head, viz., the oxidation of the organic m atter and the erosion of the surface soil
1 Tw o 3-acre fields.
2 A fo rtie th -a cre plot.
3 B ad ly in ju red b y ru s t.
b y the wind. T he deeper the layer of soil is, the lower is the content of humus and nitrogen; accord
ingly, erosion b y the wind w ill have an effect upon the humus and nitrogen content similar to that of cu ltiv a tion and m ay seriously affect the value of any such soil study as this.
The seriousness of the nitrogen problem for the semi-arid region as a whole is eviden t when it is con
sidered th at the loss in 25 years a t Indian H ead has been alm ost equal to the total percentage in the virgin soils of northwestern T exas and to much more than th at in the virgin soils of Southern Arizona. F urther, the use of a theoretically good rotation w ith legu
minous green manures over a period of nine years has failed to have an y appreciable effect in checking the rapid loss of soil nitrogen, w hile it has greatly lessened the yield of grain.
SUM M ARY.
In a comparison of 22 rotation plots no d istin ct relation has been found between the com position of the soil and the nature of the rotation. In a long cu ltivated field the till was found poorer in humus, nitrogen and organic carbon than the lacustral clay.
T h e am ounts of the above three constituents found in an y of the plots depend more upon the relative proportions of the two types of soil occurring on the plot than upon the previous treatm ent.
The longer the fields have been kep t in grasses mown for hay, the less has been the change in com position of the soil. Continuous bare cu ltivation along tree rows has caused greater losses than the
■alternation of fallow and crop in the adjacent fields.
T h e extrem e loss of nitrogen, humus and organic carbon in 25 years is about one-third of the am ounts originally present in the prairie. T his loss of nitrogen is not sufficient, however, to cause nitrogen to h ave becom e a lim iting factor in the yield of w heat on fallow ed land a t Indian Head.
A tte m p ts to substitute for the bare fallow various legum inous crops, which have been plow ed under, have decreased the yields of wheat. T h e lowered yields are evid en tly due to the drying out of the soil by the legum inous crop. In the exceptional years when the rainfall of autum n has been v e ry h eavy the yields after legum es have been as high as after fallow .
La b o r a t o r y o f Ag r i c u l t u r a l Ch e m i s t r y, Un i v e r s i t y o f Ne b r a s k a,
Li n c o l n, Ne b r a s k a.
NITROGEN AND CARBON IN THE VIRGIN AND