Proc. 12th and 13th Ann. Meeting, Soc. Prom. Agric. Sci. 1892

A PHYSIOLOGICAL BASIS FOR THE COMPARISON OF POTATO PRODUCTION

J. C. ARTHUR

THE potato crop is an important one, and much attention has naturally been given to its cultivation. The methods and various details by which the yield may be increased have been the subjects of careful inquiry by cultivators and trained observers. From the earliest times improved varieties and vigorous strains have been sought by every means at command.

In the improvement of varieties or methods of culture it is necessary to compare the yield of one variety with another, or of one method with another, and it is highly desirable that a rational basis of comparison should be fully recognized and applied. The problem is a very complex one, and includes some plain and many obscure factors.

It is the purpose of this paper to call attention to some of the factors in the comparison of productiveness, which have been overlooked or underrated, and to some extent to put this kind of experimental work upon a more rational and satisfactory footing.

The first proposition which my studies lead me to make, is that whatever increases the rate of growth at the beginning increases the yield.

There are many ways in which a difference in the rate of early growth may be brought about. Some of them, such as differences in the date of planting, of natural moisture and of fertility of the soil, are so obvious that nothing more need be said about them. Every cultivator recognizes that, if yields are to be compared, the planting must be done at the same date, upon soil of the utmost uniformity, and that the plants must afterwards receive like treatment.

But if all the obvious conditions for uniformity are complied with, the proposition will still be found to be true, as shown by the following results (Table I) selected from quite a number of experiments, all having the same bearing. In these tests, as in all those which follow, *every precaution was taken to have the treatment identical for all the plants.

* The experiments, upon which this paper is based, were carried out with every attention to detail which is advocated in the following pages. The potatoes were planted in hills three feet apart each way and cultivated both ways. The ground was of more than usually even quality, not having been fertilized is any way for several years; but to guard to the utmost against variations of soil the planting was done so that comparisons could be instituted between alternating rows. Only perfect tubers of regular shape were used; and these were assorted into lots, each lot having a variation in weight of only half an ounce, except when otherwise stated, as in the case or tubers above 8 ozs., of which the supply was limited. All cut tubers had each piece carefully trimmed until it weighed just 20 grams, or whatever other weight may have been chosen in special cases. The field was gone over daily and the date of appearance of the first shoot shove ground is each hill duly recorded. The height and number of stalks were takes at one time, early in July. When dug, the tubers of each hill were counted and weighed separately, and the tubers above 50 grams in weight (the separation being made with a balance) were also counted and weighed for the merchantable part. Owing to the care with which every operation was carried out there were no missing hills. The record of every hill was kept independently, and the summing up and averaging was not done for months after the last figure in the record was made. The full data from which the illustrative tables in this paper were selected, can be found in the manuscript records of the botanical department of the Indiana Experiment Station, at Lafayette. Ind., as follows: Table I is from the record of experiments 60 and 76; II, from Exp. 91; III from Exp. 58; IV, from Exp. 61 and 109; V, from Exp. 76; VI, from Exp. 56; VII, from Exp. 108 and 124; VIII, from Exp. 125, rows 154-158; IX, from Exp. 57; X, from Exp. 63; XI, from Exp. 60; XII. from Exp. 76. Experiments numbered below 100 were carried out in 1890; those above 100, in 1891.

TABLE I.
PRODUCT OF EARLY AND LATE VEGETATION FROM SAME PLANTING.

No. of Hills Part of Tuber
Planted
Date of
planting
Appeared
above ground
Average height
of vines, in inches
Merchantable per
hill, in grams
Total yield per
hill, in grams
a 27 Middle pieces Apr 30   16-18 days 16.8 263 366
34 Apr 30   20-26 " 16.4 230 311
b 10 Apical pieces Apr 30   15-18 " 17.7 262 375
10 Apr 30   19-22 " 17.5 257 388
c 16 Whole tubers May 3   16-17 " 22. 456 690
15 May 3   19-20 " 19. 265 492

The resulting uniform advantage of the hills first appearing above ground, although for the most part having a start of only a day or two, and the extremes being only four to ten days, is very notable. The date of appearing above ground has been taken as the most convenient index to the rate of early growth, and the data show that the average height of the plants and the yield of tubers both correspond with it.

Leaving both the too obvious and the too obscure reasons of unequal growth, we pass to several demonstrable causes of acceleration, which are generally overlooked, and which it is the purpose of this paper to more particularly deal with.

The rate of early growth, and the consequent increase in the harvest, may be promoted by wilting the tubers a certain amount before planting. This may be shown by the following data:

TABLE II.
ACCELERATING GROWTH BY WILTING THE SEED TUBERS.

  Not wilted Wilted 1 day Wilted 4 days
Number of hills planted 20 20 20
Percentage of moisture lost by wilting 0 16 37
Percentage of hills above ground in 15 days 60 100 20
Av. wt. of tubers per hill in grams, all sizes 269 344 128
Av. wt. of tubers per hill in grams, merchantable 196 266 105

The experiment just detailed shows that tubers cut into pieces and exposed in a warm room for a time before planting came up much sooner and gave a considerably larger yield than tubers freshly cut, when the wilting was not carried too far. There may be a doubt entertained, however, whether the wilting in this experiment is the true explanation of the increase; possibly the removal into the warmth and light of a living room may quicken the chemical changes in the tuber and have more to do with the subsequent rapid growth than the loss of moisture. To test this point two sets of cut tubers were placed under bell-jars standing side by side. Under one jar were also placed shallow dishes containing water, under the other jar similar dishes containing concentrated sulphuric acid, thus converting it into a temporary desiccator. The two sets of tubers were now under exactly the same treatment except that one was in a moist atmosphere and the other in a dry atmosphere. After ten days both sets were planted, one having lost 18 per cent. of moisture and the other practically none. The results, given in the accompanying table (III), show that the vegetation was much more rapid from the wilted set than from the unwilted, and that the yield was correspondingly larger. In this experiment the differences must be directly ascribed to the wilting.

TABLE III.
WILTING SEED TUBERS AT TEMPERATURE OF THE UNWILTED TUBERS.

  Unwilted Wilted
Number of hills planted 20 20
Percentage of plants above ground in 20 days 10 100
Average height of plants in inches 16.9 17.45
Av. wt. of tubers per hill in grams, all sizes 375 409
Av. wt. of tubers per hill in grams, merchantable 239 260

*For the literature of the subject, see: Sachs, Handb. d. Experimental-Physiologie der Pflanzen, 1865, p. 241; Nobbe. Versuchs-Stationen, xi, 1567, p. 218; Ueber das Anwelken de Saatkartoftel, chemnitz, 1869; Feast, Studien on dec Kartoffelknolle, Gottingen, 1S73. p. 35; Die Kartoffel als Saatgut, Berlin, 1878, p. 4$; Boree, Jont's Bot. Jahresb., vi, 1878, pt. i, p. 617; Giersberg, Der rationelle Kartoffelbau, 1878. p. 64; Keaon, Wollny's Forsch. Agrik. Phys., iii. 1880, p. 252; Goff, 5th An. Rep N. Y. Exper. Sta., 1886, p. 153; name, 6th Rep. 1887, p. 87; Leydhecker and Mann, Wollny's Forsch. Agrik. Phys., x, 5857. p. 239; Green, 7th An. Rep. Ohio Exper. Station, 1888, p. 526; Speth, Bull, Ga. Exper. Station, No. 8, 1890, p. 123.
See also: Muller-Thurgau, Just's Bot. Jahresb., xv, 1887 pt. ii, p. 335.
From the above data, and much more which cannot be conveniently given here, the conclusion has been reached that the length of time taken to wilt the seed tubers is not material, but that the percentage of moisture evaporated cannot, as a rule, exceed 20 per cent, without injury. But within that limit the greater the loss of moisture before planting the better the tuber grows, and the larger the resulting crop.*

It has been found that placing young or freshly dug potatoes in a cold chamber for a few days, then planting them, will produce a remarkable decrease in the time required for them to sprout and show above ground, and a correspondingly greater after growth of foliage and tubers. My trials have been made with a cold chamber standing at 10° C., and with uniform results, but have not been extensive enough to present the data in the form of statistical tables.† Whether such treatment will accelerate growth of seed material after a winter's rest is doubtful.

The rate of early growth may be increased by a more ready access of moisture to the interior of the planted tuber. This statement may be tested in several ways. Probably one of the most satisfactory methods is to plant whole potatoes, which are alike in every respect except that one set has the thin impervious skin of the tubers removed, without, however, disturbing the eyes. In a trial of this kind the results, given in the accompanying table (IV), were uniformly in favor of the decorticated tubers, both in early growth and final yield.

TABLE IV.
SEED TUBERS WITH AND WITHOUT CORTEX.

  1890 1891
Tuber
intact
Tuber
decorticated
Tuber
intact
Tuber
decorticated
Number of hills planted  40 40 40 40
Percentage of hills above ground in 22 days 35 85 68 98
Av. Number stalks  per hill 5 7 6 6
Av. number tubers per hill 13 16 15 17
Av. wt. of tubers per hill in grams, all sizes 493 623 842 1053
Av. wt. of tubers per hill in grams, merch'table 249 343 619 832

Another method of testing the matter is by cutting open part of the tubers, which permits the moisture of the soil to come in direct contact with the interior of the tubers. A trial of this sort (Table V) in which the control hills were planted with whole tubers, and the hills for comparison with tubers divided into a certain number of pieces by transverse cuts just before covering, gave an advantage in early growth for the hills with divided tubers, a gain which must be chiefly or wholly ascribed to the more ready absorbtion of moisture by the seed tubers. The subsequent development, however, was more strongly influenced by other factors which the treatment introduced, particularly the increased number of stalks per hill and any advantage received from the earlier start was quite obscured and overcome by the later influences.

TABLE V.
DIVIDING SEED TUBERS BEFORE PLANTING.

  4 oz. Tubers. 5 oz. Tubers.
  Undivided Divided into
four pieces
Undivided Divided into
six pieces
Number of hills planted 40 40 40 40
Percentage of hills above ground in 18 days 43 85 63 90
Av. number of stalks per hill 8 11 9 13
Av. wt. of tubers per hill in grams, all sizes 570 569 576 600
Av. wt. of tubers per hill in grams, merchantable 344 292 350 258

In selecting tubers for comparison so apparently insignificant a difference as the degree of natural roughness of the surface is important. The roughness due to scab or other diseases, or to the action of worms, mice, etc., or to other extraneous causes, is not referred to, but the finely checked and loosened surface of well grown tubers due to the partial separation of the outer part of the skin.

TABLE VI.
ROUGH AND SMOOTH SEED TUBERS.

  Smooth Rough
Number of hills planted 60 60
Percentage of hills above ground in 22 days 45 57
Av. number of stalks per hill 3.4 3.6
Av. weight of tubers per hill in grams, all sizes 432 462
Av. weight of tubers per hill in grams, merchantable 304 321

The tubers selected to test this statement were from the same strain, and in all known respects were alike, except in the character of the surface. The results show a very marked and uniform advantage for the rough skinned tubers over the smooth ones. And here again we find that the superiority is first manifested in the earlier appearance of the plants above ground.

*See Wollny & Pott, Just's Bot.
Jahresb., iv, 1876, p. 1194.

The cause of this difference in vigor does not lie, of course, in the skin of the tuber, which only indicates some variation of a more profound nature. It has been ascertained that the specific gravity and the percentage of starch increase with the degree of roughness. The rough tuber, therefore, is much richer in reserve food for the forthcoming plant, than the smooth tuber of the same size, which may in part account for the difference in behavior.*

Every potato tuber has a number of eyes, each one holding the possibility of a shoot, which in time may return to the cultivator from one to six or eight new tubers, as good as the parent. But there are more eyes on a potato than the cultivator knows what to do with, and especially is there an inextricable jumble of eyes at one end of the tuber. The stare of these numerous eyes has been the nightmare of the experimenter since the early part of the century, increasing in an arithmetical intensity as the years have passed.

"Man entweder die kleine Brut darzu nimt, oder die grossen Aepfel in vier Theile zerschneidet." Hoppe, Kurzer Bericht von den Erd-Aepfeln, 1747, p. 18.

The oldest work upon the cultivation and value of the potato is a thirty-two page pamphlet by Tobias Conrad Hoppe, dated 1747. The author says in quaint German† that the seeds are not used for planting, but either small tubers are taken, or else large ones cut into four parts. This early German method continued to be the general practice for nearly a hundred years.

It could not fail to come to the notice of cultivators after a time that the planting of large potatoes gave a better yield than small ones, but it was not till near the beginning of the second quarter of the present century that the fact was definitely formulated.

It was about this time that careful experiments upon the treatment of seed material were first undertaken, and at once led to the now thoroughly familiar conclusion that "the larger the seed planted the greater the yield." To quarter the large tubers did not long suffice, and the trials extended into a series, especially within the last twenty-five years, of cutting to one eye, two eyes, three eyes; seed ends, stem ends, middles; quarter tubers, half tubers, whole tubers, with all manner of variations. There appears to be a fascination in this kind of research, for it has been tried by nearly every experiment station in the country, and by some repeatedly, while the few remaining stations will, doubtless, soon follow. It has also been tried by many private individuals, and their conclusions recorded in the agricultural press. Each experimenter usually carries out and reports his work as if it were the first trial ever made.

* The following references give some of the most important experiments leading to this conclusion: Putsche & Bertuch, Versuch einer Monographie der Kartoffeln, Weimer, 1819. p. 09; Lindley, Trans. Roy. Hort. Soc., 2nd ser., i. pp. 153, 445; Hexamer, Amer. 1866, p. 98; Maw. Jour. Roy. Ag. Soc., 1867. p. 554; Farquhar. Rep. 17.5. Dep. Agric., 1867. p. 187: Sorauer, Neue Fuehling'sche Iandw. Zeitung, 1871; Hellriegel, Amtl. Vereinsblatt d landw. Centraly, d. . Brandenburg, 1872; Drechsler, Oester. tandw. Wochenblatt, 1875 and 1876; Same, Journ. f, Landw., 1876; Some, Just's Bot. Jahresb., vi, 1878, p. 615; Voesler, Wochebl. f. Land- und Forstwirthschaft, 1876; Sanborn, Rep. N. H. Bd. Agric., x, 1880, p. 279; Same, Amer. Agric., xli, 1882, p. 873; Lozenby. Annual Rep. Cornell Univ. Exper. St. i,1880, p. 105; Same, Annual Rep. Ohio Exper. St., i, 1882. p. 51; ii, 1883; iii, 1884, p. 91; iv, 1885, p. 70; Sturtevant, Annual Rep. N. Y. Exper. St., ii, 1994, . 117; iii, 1885, p.79; iv, 1886. p. 47; Green, Annual Rep. Ohio Exper. St., v, 1886, p. 154; vi, 1887. p. 206; Krensler. Landw. Jahrb., 1886, p. 309; Scovell, Bull. Ky. Exper. St., No. 16, 1888, p. 9; Alvord, Bull. Md. Exper. St., No. 2, 1888, p. 21; Bishop, same, p. 27; Emery, Annual Rep. N. Y. Exper. St., viii, 1889, p. 232; ix, 1890, p. 375; Kinney, Annual Rep. R.I. Exper. St., 1890, p. 109; Brunk, Annual Rep. Md. Exper. St., iii, 1890, p. 108; Carman, The New Potato Culture, 1891, p. 110.
Plumb, Bull. Exper. Sta. of Tenn., vol. iii, 1890, p. 2.
Maw, Proc. Royal Agric. Soc., 1867, p. 552.

So much labor ought to show an advancement in knowledge, and yet we find that the conclusion of the last experimenter is in the main the same as that of the first, i.e., "the larger the seed planted the greater the yield,"* or else, as in the tests of stem ends, seed ends and middles, the results are unexpectedly variable. In many cases the experimenter has, apparently, not risen above empiricism, and possesses no theoretical guide based upon the underlying principles of structure and growth, by which to plan his work or to test the results.

But few experimenters have interpreted the dictum that "the larger the seed planted the greater the yield" with much exactness. An accurate measurement of the size of the seed material is not practical, and the assortment of the tubers has usually been into large, medium and small, and when cut into pieces, unless halved or quartered, they have been rated according to the number of eyes the piece possessed. A few experiments reported by station workers have been conducted with weighed material, and among the best of these is one by Plumb,† conducted in Tennessee, in which eight sizes of whole tubers were used, ranging from 1 to 14 ounces, a variation of two ounces being permitted for each size. The result was a uniform increase in the number of tubers and in the total weight of the product from the smallest to the largest size. A carefully conducted experiment carried out in England twenty-five years ago apparently did better than this, however. It was found‡ that tubers selected to weigh 1, 2, 4, 6 and 8 ounces gave results increasing regularly with the increase in weight of the seed material. My own experiments show that sizes differing but an ounce from one another and with a variation for each size of only half an ounce, gave results uniform with the above. From these and other data that might be cited, the deduction may be safely made that the product varies as the weight of the seed material.

Considering this proposition as fundamental, we are in a position to test the various questions regarding the several methods of cutting tubers.

The following references include some of the best recorded experiments in this line: Green, Annual Rep. Ohio Exper. Sta., v. 1886, p. 154; vi, 1887, p. 206; Emery, Annual Rep. N. Y. Exper. Sta., viii, 1889, p. 225; Kinney, Annual Rep. R. I. Exper. Sta., iii, p. 114.

Starting with the use of 1, 2 and 3-eye pieces,§ the method can be shown to be wholly unsatisfactory, as usually carried out, not only because no account is taken of the weight of the individual pieces, but because of the unwarranted assumption that the number of eyes on a piece holds some definite relation to the number of shoots or aerial parts produced. The following tabulation (VII), the result of trials made the present season, indicates that the number of stalks is but slightly increased for the same weight of tuber even when the number of eyes increases two, three or even a hundred times; and as it can furthermore he shown that the number of tubers in a hill is within certain limits determined by the number of stalks (See Table VIII), it is evident that the number of eyes on a tuber, or part of tuber, is quite an insignificant factor compared with the weight of the piece.

TABLE VII.
NUMBER OF STALKS FROM SEED MATERIAL WITH DIFFERENT NUMBER OF EYES.

20 Whole 40 gram tubers with 226 eyes gave 104 stalks
20 " " " " 162 " 100 "
20 Cut " pieces " 60 " 102 "
20 " " " " 40 " 93 "
20 " " " " 20 " 83 "

TABLE VIII.
RELATION OF NUMBER OF TUBERS TO NUMBER OF STALKS.

1 Hill Contained 2 Stalks, each with 4.0 Tubers per Stalk.
2 " 3 " 2.7 "
7 " 4 " 2.4 "
6 " 5 " 2.5 "
4 " 6 " 2 "
12 " 7 " 1.6 "
18 " 8 " 1.5 "
12 " 9 " 1.6 "
7 " 10 " 1.4 "
7 " 11 " 1.4 "
5 " 12 " 1.3 "
13 " 13 " 1.3 "
3 " 14 " 1.1 "
1 " 15 " 1.1 "
2 " 16 " 1.3 "
100 hills, averaged 9 stalks each, with 1.5 tubers per stalk.

*Some of the results obtained by other experimenters in the use of "seed ends, stem ends and middles" may be consulted as follows: Franz, Studien an der Kartoffelknolle, 1873. p. 10; Same, Die Kartoffel als Saatgut, 1878. p. 108; Werner Kartoffelban, 1876. p. 50; Sanborn, Rep. N. H. Bd. Agric., x, 1880, p.288; berne, Amer. Agrlc.,xl,, 1882,p. 125; Sturtevant, Annual Rep. N. Y. Exper. St. ii, 11584. p. 112; iii, 1885, p. 68; iv, 1886, p. 48; Goff, Same, iv. I88, pp. 202, 208; Lazenby, Annual Rep. Ohio Exper. St., iv, 1885, p. 72; Hence Rep. Mass. Bd Agnie. 1805, p. 14; same, 1886, p. 113; Taft, Bull. Mich. Exper. St. No. 70, 1891, p. 23; Carrmen, The New Potato Culture, 1891, p. 112.
Passing now to a comparison of the terminal and lateral eyes (or, in the language of the farmer, seed ends and middles), a single table (IX) will suffice to show the character of the available statistical data when using pieces of uniform weight cut from tubers also of uniform weight. It is evident from the experiment that the terminal pieces in this case were much superior to the lateral pieces in rate of early growth and also in subsequent growth and yield ; and other trials uniformly support these conclusions.*

TABLE IX.
COMPARISON OF TERMINAL AND LATERAL GROWTH.

  Apex Side
Number of hills planted 40 40
Percentage of hills above ground in 21 days 65 43
Av. number of stalks in a hill 4.6 4
Av. height in inches 15.7 14.4
Av. number of tubers per hill, all sizes 12 9
Av. number of tubers per hill merchantable 4 8
Av. weight of tubers per hill in grams, all sizes 534 395
Av. weight of tubers per hill in grams, merchantable 358 265

When we seek for an explanation of the differences in behavior of the terminal and lateral parts of a tuber, we turn at once, as in the case of rough and smooth tubers, to the examination of the reserve material, especially of starch. The question seems to be immediately answered when we find that the amount of starch is several per cent. larger in the terminal than in the lateral region of the potato, the difference corresponding with the difference between the percentage of starch in rough and smooth tubers.

But a closer scrutiny will show that, although this explanation is undoubtedly correct so far as it goes, yet it is not adequate, for we find that the gain in yield of rough tubers over smooth ones amounted only to 7 per cent., while with practically the same variation in starch content, the terminal portion of the tuber gives 35 per cent. increase in yield over the lateral portion. Although these figures might not hold good in all cases, yet they may be taken as indicating that there is a greater difference between the terminal and lateral portions of a potato than can be accounted for by the difference in the percentage of starch.

To understand the solution of the problem it will be necessary to briefly examine into the morphological and biological nature of the potato tuber. It is of fundamental importance to know that a potato is morphologically a much thickened stem, and that the eyes are buds. Biologically, a potato increases in length by apical growth, as do other stems, and the eyes at the base, or stem end of the potato are the oldest, and they decrease in age toward the apex of the tuber, or the seed end, the terminal eye being the youngest—the last to mature.

*The following may be consulted for some further account of structure and physiology: Franz, Jour. f. Landw., 1873, p. 311; Bishop, Annual Rep. Md. Exper. St., i, 1888, p. 69; Arthur, Bull, Ind. Exper. St., No. 15, 1888.

Instead of giving a detailed explanation of the structure and development of the potato tuber,* it will suffice for the purposes of this article to illustrate the same by reference to the well known characters of aerial stems. It must be remembered that a ripe potato tuber is in a resting stage, and therefore corresponds to the winter conditions of ordinary stems. If we take the end of a branch in midwinter, showing the last season's growth of a maple, hickory or other deciduous tree or shrub, we shall have a fair illustration of the parts of a potato.

Everyone knows that the terminal bud of such a branch will, at the proper time, give the most vigorous growth, and that the lateral buds will show less vigor in proportion to the distance they are removed from the apex. The number of branches at a node is not limited to the growth of a single conspicuous bud, but under certain conditions, such as an injury to the main bud, may be multiplied indefinitely by the formation of accessory or adventitious buds. The reasons underlying these facts, while matters of interesting and profitable inquiry, may be ignored in this connection, it being sufficient to have the facts recognized.

In the potato the buds are sunken in the tissue of the tuber, and possess no special protective coverings, and hence display no visible differences between the strong buds and the weak ones such as everyone recognizes in tree shoots. These differences do, however, exist. Reverting now to our experiment with terminal and lateral eyes, we may explain the difference between the results in an intelligent way by saying that it is due in part to the larger amount of reserve food material in the form of starch in the apical portion of the tuber, but more especially to the greater vigor of the terminal eyes.

Theoretically the vigor of the eyes decreases in a definite series from the terminal to the furthermost lateral (basal) eye. In a field test it is difficult to show this, owing to the accidental variations due to causes which the experimenter cannot control, and which usually exceed or quite obscure the small differences due to position of adjacent eyes. An approximate demonstration, however, can be made. The accompanying table (X) shows the data obtained by planting pieces of equal weight from four regions of the potato, in which the results accord fairly well with theoretical requirements.

TABLE X.
POSITION OF THE EYES IN RELATION TO VIGOR.

  Apex Near
Apex
Near
Base
Base
Number of hills planted 40 40 40 40
Percentage of hills above ground in 22 days 90 87 77 52
Av. number of stalks in a hill 2.92 2.55 2.07 2
Av. height in inches 14.35 13.17 13.22 12.6
Av. number of tubers per hill, all sizes 5.95 5.75 5 4.6
Av. wt. of tubers per hill in grams, all sizes 294 281 268 226

The careful observer of woody plants has not failed to notice that not only is there a difference between the vigor of the buds upon the same shoot, but that of two shoots the larger one will have the stronger terminal bud and push into more vigorous growth. We have already found that large tubers in like manner show a better growth than small ones, which has been accounted for by a difference in the amount of available food material held for the plant. Is it not possible, however, that part of this difference is due to the comparative strength or vigor of the eyes, apart from the question of the amount of reserve food? The problem can evidently be put to the test by planting equal weight pieces from different size tubers. This is carried out in the accompanying table (XI), using the terminal eyes from five sizes of tubers ranging from one to nine ounces respectively. The results show an almost uniform increase from the smallest to the largest tubers, although the pieces planted were all of the same weight. Here again we also see that there is a relation between the size of the crop and the rate of early growth.

TABLE XI.
VIGOR OF TERMINAL EYES, USING FIVE SIZES OF POTATOES.

  1-1 1/2 oz. 2 1/2-3 oz. 3 3/4-4 1/4 oz. 5 3/4-6 1/4 oz. 8-9 oz.
Number of hills planted 20 20 20 20 20
Percentage of hills above ground in 18 days 15 35 40 50 50
Av. height in inches 16.1 16.4 16.9 I7.6 17.9
Av. number of tubers per hill, all sizes 9.5 9.9 10.1 8.8 10.6
Av. number of tubers per hill, merchantable 1.8 2.3 2.6 2.7 3.1
Av. wt. of tubers per hill in grams, all sizes 317 352 372 382 419
Av. wt. of tubers per hill in grams, merchantable 134 182 198 254 256

The proposition with which we started, that increase in the rate of early growth of the potato plant increases the yield, has now been demonstrated to be true, and to be brought about by a variety of causes, viz.:

a. Wilting of the seed material. d. Roughness of the seed material.
b. Access of moisture to seed material. e. Weight of the seed material.
c. Cooling of the seed material. f. Vigor of the seed material.

There are other causes which may also operate in the same way, such as the temperature of the storage room during winter, starting the tubers in a light room before planting, removing sprouts before planting, etc., but these may be passed without further mention, as in the main sufficiently obvious.

The demonstration of a biological problem always rests upon the assumption that all other factors besides those mentioned are equal. Important factors may, however, be unwittingly overlooked, it has already been shown that the exact weight of the seed material is of the utmost importance, although often ignored, and it now remains to point out that the distribution of space is equally important. The necessity of having hills and rows spaced alike, when yields are to be compared, is, perhaps, sufficiently obvious to every one, and may be passed without comment, but a less evident cause of irregularity needs attention.

The potato plant is a gross feeder, and takes its nutriment from a very restricted area. It is consequently easy to crowd a hill or row with more plants than the soil will properly support. When this is done the plants are smaller and the yield is less, whatever the promise of the early growth may have been. Here lies the explanation of the common belief that planting large whole potatoes increases the number of small tubers in the crop. The heavy seed material starts more shoots, and consequently more tubers in the hill, than the roots in their crowded condition can find nutriment for. An unusual proportion of the tubers, therefore, fail to reach full size.

TABLE XII.
DIVIDED TUBERS COMPARED WITH WHOLE TUBERS.

  1 1/2-2 oz. 3 3/4-4 1/4 oz. 4 3/4-5 1/4 oz.
Undivided Divided into
two pieces
Undivided Divided into
four pieces
Undivided Divided into
six pieces
Number of hills planted 40 40 40 40 40 40
Percentage of hills above ground in 18 days 70 68 43 85 63 90
Av. number of stalks per hill 4.2 5.6 8. 11.4 8.8 13.8
Av. number of tubers per hill, all sizes 8. 8. 4 13.0 16. 13.8 17.9
Av. number of tubers per hill, merchantable 2.5 3. 4.2 3.8 4.3 3.5
Av. number of tubers per stalk, all sizes.0 1.9 1.5 1.7 1.4 1.6 1.8
Av. wt. of tubers per hill in grams, all sizes 343 372 570 569 576 600
As. wt. of tubers per hill in grams, merchantable 237 266 344 292 350 258

* For further data see: Goff. Annual Rep. N, Y. Exper. St., v, 1886, p. 151; vi, 1887, p. 86; vii, 1888, p. 162; Emery, same, viii, 1889, p. 236.
An error is sometimes made by supposing that a whole tuber may be represented quite as well by two small ones, or two pieces, which taken together make the same weight. The following table (XII) shows the results obtained by comparing one tuber in a hill used whole, with one tuber in a hill cut into pieces of the customary size, and the parts slightly separated from one another in the hill. The comparisons are carried out with three sizes of tubers. It will be seen that dividing the tuber into separate parts increased the number of stalks in a hill, and decreased the number of tubers per stalk. An average of more than six stalks in a hill materially interfered with the growth of the tubers and lessened the yield of the merchantable product. The table shows clearly that multiplying the pieces in a hill, even when the total weight remains the same, increases the number of plants grown upon the same area of soil, and when this increase exceeds a certain rather low optimum (six stalks to the hill in the experiment cited) the development of the tubers is materially restricted. Even when the weight of seed material planted is so small that there is ample room in the hill for development, as in the case of the 2 oz. tubers, the experiment shows that varying the number of pieces of seed material varies the product.*

CONCLUSIONS.

Physiological.—The potato tuber is physiologically the resting state of a branch having a number of buds. Each tuber possesses a certain amount of reserve force, partly in the form of starch, and partly in other forms designated as vigor. The reserve force of the tuber is greatest at the apical (seed) end and becomes less toward the base; it increases with the size of the tuber and with its natural roughness; and it may be made more available by wilting the tuber, exposing it to cold, or facilitating the absorption of moisture. Dividing the tuber divides the reserve force, and increases the number of lines in which it is distributed, or in other words, the number of plants to which it gives rise.

Practical.—From these physiological considerations the following fundamental propositions are deduced for use in comparison of potato production:

  1. The seed material must be of the same weight, roughness and number of pieces.
  2. If the tuber is divided, only the same regions of the same weight tubers are comparable.
  3. All other factors are understood to be equal, except the one for which the comparison is made.