Further Tests with Black Walnut Varieties L. H. MacDaniels and J. E. Wilde , Cornell University

In 1937 the Northern Nut Growers Association committee on varieties and judging standards proposed a tentative schedule for the judging and evaluation of black walnut varieties(1). It was pointed out at that time that for one reason or another none of the schedules which had been used in judging walnuts were satisfactory and usable in giving an accurate estimate of the cracking quality and value of a variety. It was recognized also that the schedule proposed was only tentative and that it would need to be modified in the light of future testing and experience. In 1939 the question was again considered(2) and on the basis of tests which had been made, changes were proposed which would make the schedule more realistic. Since then many tests have been made using the modified schedule. The purpose of this paper[A] is to give the data secured in these tests and to consider again the value of the schedule and possibilities of improvement.

Recently a number of papers have been published dealing with the evaluation of black walnut varieties. In 1941 Kline and Chase(3) compiled the available published data and additional tests made by the Tennessee Valley Authority on nut weight and kernel percentage of black walnut selections. Two hundred and twelve clones and 335 tests are reported. As would be expected the samples of the same variety from different localities show variation in weight per nut and in total per cent kernel. For example, in 12 samples of the variety Ohio the weight per nut varies from 14.8 grams to 18.7 and the per cent kernel from 16.6 to 32.9. Twenty-one tests of Thomas show variations in single nut weight from 16.7 to 25.0 grams and in per cent kernel from 19.0 to 30.0. In general the samples grown in the north were made up of smaller nuts with less per cent kernel, indicating that the varieties were not suited to that latitude.

In 1942 Kline(4) worked out a somewhat technical method of evaluating walnut varieties on the basis of cash return per hour of labor spent in cracking with a hand operated cracker. A formula is proposed in which the variables of price and other factors may be substituted. The approach is on a commercial basis and the method is not intended for use in evaluating small samples. The paper represents many tests and establishes or affirms by statistically treated data several points of general interest in walnut testing, namely, (1) that a 25 nut sample is large enough to show varietal or other differences of a gram in total weight or 1 per cent of kernel weight, (2) that unless extreme accuracy is desired, moisture content may be ignored in making tests of 25 nut samples if the nuts have been hulled and air dried for about two months and (3) that the mean weight per nut and per cent kernel of nuts from the same tree may vary appreciably from year to year, for example a variation of 4.9 grams per nut and 3.3 per cent in kernel weight is reported for Snyder. Such variation is recognized and emphasizes the necessity of testing a variety in any locality for a number of years if correct valuation is to be made.

In Kline's paper earnings per hour for fifteen black walnut selections are given showing a maximum of $0.279 for the variety Norris, $0.245 for Ohio down to $0.12 for an unnamed seedling.

Lounsberry(5) published kernel cavity measurements for 64 clonal selections and related these to kernel weight per nut. Measurements of the thickness of the partition separating the halves of the kernel are also given. He does not relate these characters to scoring or cracking quality.

The purpose of the scoring system under discussion in this paper is to provide a realistic method of judging the relative merit of different clones of black walnuts that can be used mostly by members of the Northern Nut Growers Association or others having some skill in cracking technique. At the present time the Association has little reliable information either as to the performance of different varieties under different conditions in any one locality, from year to year on the same tree, or the suitability of any one variety growing in far different parts of the United States. It is important that such information be available and a workable basis of evaluation would be of the greatest value in obtaining it. Much of our information at the present time is from the many tests made by N. F. Drake(6, 7, 8) which are of great value in rating varieties. His schedule is an improvement over any previously proposed but fails to provide standard sampling and cracking procedure and includes the items of flavor and color which are in no way objective characters. The use of a point score based on the concept of a "perfect nut" is cumbersome and considered undesirable by the committee.

It is recognized that the value of a variety depends also upon the bearing habit of the tree, the nature of the husk, disease resistance and other characters.

It has been five years since the present schedule was proposed and enough tests have been made to give a basis for judgment as to the merits and weaknesses of the schedule. As stated in the original committee report it is generally agreed that the best measure of the value of a nut of any clone is the amount of usable or marketable kernels that can be obtained from a given weight of shucked nuts with the least labor. The characteristics of the nuts that contribute to this value are recognized as follows:

1. The size of the individual nut.
2. The per cent of kernel of total sample weight recovered without recracking
and without the use of a pick.
3. The total per cent of kernel of total weight of sample.
4. The number of quarters.
5. The plumpness of the kernels.
6. The number of empty nuts or nuts with shrivelled kernels in the sample.

Flavor and color may be important but are so dependent upon personal preference and on the treatment of the samples before testing that they cannot be rated numerically.

In considering the value of any schedule the following questions are pertinent:

Table 1 gives data on replicate samples tested by the same operator. In the samples of Spear, numbers 1-6 the variation is as follows: weight of single nut 1.3 grams, per cent kernel first crack 2.9, total per cent kernel 2.6, number of quarters 3, penalties 4.5 points, score 9.2 points. In scores figured without penalty the variation is 5.4 points. Sample No. 7 was cracked November 4 before the nuts were dry and hence is not comparable with others.

Analysis of these differences indicates that the variation in nut weight is closely related to the number of shrunken and empty nuts in the sample. This is a difficult factor to evaluate in a practical way. At the time of the 1939 report it was suggested that the score should be figured on the basis of filled nuts. This cannot be arranged easily in testing because if the operator cracks the nuts before weighing there is almost sure to be loss of fragments of shell. Trying to correct the original weight in any way is necessarily inaccurate. Deciding whether or not the kernel of a nut is sufficiently shrivelled to deserve a penalty is a matter of judgment which is a personal matter.

The variation in per cent kernel first crack and total per cent kernel probably represents fairly the difference in the samples. The total per cent is a wholly objective value and varies practically as much as the per cent first crack. Uniformity in the number of quarters is striking. This large number is undoubtedly related to the fact that many of the kernels were shrunken enough to be penalized and others were perhaps shrunken enough so that they did not tightly fill the shell cavity. In general it may be said that the more tightly the kernels fill the shell the more difficult it is to extract large pieces. Thus having the kernels a little shrunken but not enough to seriously reduce their weight favors a higher score. Of course, in some varieties the kernels may he plump and still not fill the shell tight enough to make cracking difficult. This is a desirable condition.

Variability in penalties is more important (i. e. 4.5 points) than any other factor in influencing the final score. Without the penalties the scores of samples 1 to 6 would be 87.5, 84.0, 83.6, 83.7, 82.1 and 82.8 respectively which is fairly uniform. Statistically the presence of empty or shrivelled nuts in a lot from which samples are taken increases the number required to make a satisfactory sample by greatly increasing the individual variation of the single nut.

TABLE 1

Variation in the score of tests of duplicate samples made by the same operators.
Twenty-five nut samples. Nuts grown at Ithaca, N. Y.

1942. Black Walnuts.

With the variety Snyder a difference of 2.4 grams in weight per nut in samples 1 to 5 suggests poor sampling technique as this is an objective value. A difference of 4.2 per cent in first crack suggests carelessness on the part of the operator in cracking or difference in soaking as this is quite out of line with the variation of .8 per cent in per cent weight of total kernel. The difference of 16 quarters is considerable but represents only 1.6 score points. As with the Spear the variation in penalty of 4 points is greater than other factors except per cent first crack (i.e. 4.2% points). The difference in score of 5.5 points is obviously greater than desirable, but probably indicates the relative value of the samples. Without penalties the difference is 4.5 points.

Sample 7 of Spear and number 6 of Snyder were cracked November 4th when only partly cured and show the importance of curing in obtaining an accurate rating for a sample. The score of each variety was increased materially in all characteristics and no shrivelling was apparent. As a practical means of recovering the kernels in large pieces, cracking before the nuts are dried out is a decided advantage provided the kernels are cured before they are stored.

The duplicate samples of Eldridge check very closely and show no significant differences.

In Table 2 are given the results of ten tests on carefully replicated random samples of Snyder black walnuts. In making these samples the nuts were spread in a single layer on the floor and lots of 25 cut off the edges of this layer without selection of any kind. Even with such selection there is a variation of 1.2 grams in the average weight of single nuts from different samples. Per cent kernel first crack shows a minimum of 21.8 and a maximum of 26.9 in the ten samples. This difference is related mostly to the presence of 3 empty nuts in the low scoring sample as compared with none in the high scoring sample. The high score is also in part due to soaking. This variability is about the same as with total per cent kernel indicating that cracking technique was uniform. Comparing samples 1 and 2 in more detail it is found that the difference of 11.6 points in the score is caused by the presence of empty nuts in the sample. The average weight of kernels per single nut in sample 1 is 4.9 grams. The difference in the weights of the kernels of the two samples is 15 grams or about the weight of the kernels of 3 nuts. These empties also reduce the score by reducing the number of quarters recovered. Where empty nuts are involved, it is doubtful if random sampling will reduce variation unless the size of the sample is greatly increased, a practice which is not a practical solution in that a 25 nut sample is about as large as can be handled with any facility. It would seem that this difference in scores was a fair indication of the merit of the two samples. The scores of the other samples show a fair degree of uniformity. The high score of sample 4 is probably related to the soaking treatment though the scores of sample 3 also soaked is lower than that of sample 6 which was not soaked. It seems that when these conditions and with this variety stored in a fairly high humidity, soaking had little effect except to increase the number of halves recovered.

TABLE 2

Cracking tests by single operator with 10 random replicate samples of
Snyder black walnuts. 1942 crop. 25 nut samples.

Another lot of 24 random replicate 25 nut samples of Ohio black walnut from the original tree was made by scooping the nuts out of a bag with a quart berry box which held about 25 nuts. Care was used not to select the samples in any way. The lightest sample 3 weighed 385 grams, the heaviest 22 weighed 434 grams or a difference of 2 grams per nut. The score of these two samples was 85.0 and 85.4 respectively apparently because there were no empty nuts in either sample.

The results of tests on 18 of these replicate samples of Ohio are given in Table 3. The nuts were apparently a uniform lot. The kernels while of good quality were in most cases not quite plump and did not fill the cavities of the shell tightly. This doubtless accounts for the large number of quarters recovered. The kernels on the whole were plumper than with the variety Snyder reported in Table 2 and there were fewer empty nuts. Of the samples that were not soaked the variation of 4.3 per cent in the per cent first crack is of the same order as variation of 3.6 per cent for total per cent kernel and indicates uniform cracking technique.

The data in Table 3 gives evidence of the effect of treatments before cracking. The first nine samples marked with an asterisk were held for several weeks in a damp cellar and have an average test score of 86.6. The last seven samples were held in a dry but unheated room for a week before cracking and show an average test score of 83.7. The average score for the two soaked samples was 93.9. Soaking also increased the number of halves and quarters recovered in the same way as shown with variety Snyder in Table 2. None of these samples was excessively dry. In this table the lowest score (sample 19) is directly related to the presence of 3 empty nuts in the sample. The low score of sample 21 is mostly related to low per cent first crack which is caused by large number of bound quarters and the high penalty related to empty nuts and shrivelled kernels. These scores seem to indicate the value of the samples but bring out the difficulty of obtaining equal scores from such replicate samples. The other scores in the table are probably as close to each other as can be expected with samples of this sort.

In this and the preceding tables the number of bound quarters is given as an indication of cracking technique. With the Hershey cracker the nuts of many varieties will split into four quarters without releasing the kernels. The number of such bound quarters is increased if the operator does not put sufficient pressure on the anvils to crush the shoulders of the nut and free the kernel. On the other hand if too much pressure is used and the anvils brought too close together the kernels will be crushed and the score affected adversely. With some varieties, for example, the Adams as shown in samples 1 and 2 in table 5, the nuts are so pointed at each end that the standard anvils do not strike the shoulders of the nut and many bound quarters result. With such varieties cracking with a hammer would probably give a better score. Anvils with deeper cavities in the ends would be an advantage for such nuts.

TABLE 3

Tests by the same operator of duplicate samples of Ohio black walnuts, treated
in various ways before cracking. 25 nut samples. 1942 crop.

TABLE 4

Variation in score of replicate samples of 3 varieties of Black Walnuts
tested by different operators and of same varieties from
different sources

Table 4 gives the results of tests of similar samples of three varieties from four different sources by three different operators. The tests are not satisfactory because pretreatment was not uniform and there is insufficient data on penalties which are omitted. Some samples of the varieties Ten Eyck and Thomas contained empty nuts and shrivelled kernels which would preclude equal scores. The variety Ohio was uniformly filled from all sources. In the variety Ten Eyck there is a difference of 10.5 per cent in total per cent kernel in samples from the Baum orchard. This was related to 6 empty nuts in the sample cracked by operator 3. In the variety Ohio in which the kernels were plump the greatest variation between duplicate samples in total per cent kernel is 3 or only about 10 per cent of average total per cent kernel.

An examination of these data show the following points of interest: (1) that the duplicate samples showed considerable variation in weight of single nut and total per cent kernel, characters not dependent on personal skill or judgment. Operator 2 did not crack the whole sample of 25 and may have selected the larger nuts, thus securing a greater weight per nut with all varieties. The superior filling of the nuts of Ohio appears to be related to the fact that in the orchards in question this variety was observed to hold its leaves longer than the others which lost their leaves in late summer before harvest by leaf blight. Shrunken kernels are a logical result of early defoliation.

In the per cent of kernel obtained in first crack operator 1 recovered a higher per cent than operator 3 in all of the eleven possible comparisons and higher than operator 2 in 9 out of 12 possible comparisons. This probably is the result of soaking the samples by operator 1 and not by the others or possibly due to greater skill or care in cracking. The number of quarters recovered by operator 1 is greater in all cases than that obtained by either operator 2 or 3. This is also a result of soaking or skill or both. The score of operator 1 was in all tests of duplicate samples higher than that obtained by operator 3 and higher than the scores of operator 2 in 9 out of 12 comparisons.

The scores of the different samples are apparently mainly determined by the per cent recovered at first crack and the number of quarters, at least the only cases where the scores of operator 2 exceed those of operator 1 are where the per cent first crack and the number of quarters are greater for operator 2. This is related to the presence of empty nuts.

The data obtained for the variety Thomas by operator 1 and 2 show for the most part the same relative scoring of samples from different sources. For example with both operators the score of the samples from the Weber orchard was lower than that from the Jones and Baum orchards and the sample from the Jones orchard scored higher than that from the Baum orchard. In the samples from the Worton orchard the relative scores are reversed. The scores o£ operator 3 are quite out of line. With the variety Ten Eyck the differences between scores of samples from different sources are not consistent. Operator 2 obtained scores that were essentially alike for all four samples whereas the scores of operator 1 show differences of more than 10 points. This is related to empty nuts in the sample. With the variety Ohio there is reasonable uniformity in the scores obtained by all operators. This was the only variety with well filled nuts and for that reason alone the score would be less variable.

TABLE 5

Tests by different operators on duplicate samples of black walnuts,
soaked and unsoaked. 25 nut samples. 1942 crop.

The average scores of all samples of each variety are Ohio 90.0, Thomas 83.4, and Ten Eyck 82.4. These are not out of line either with the scores obtained for these varieties elsewhere or the relative merit of the varieties.

Because of the variability obtained in the tests shown in Table 4, another series of tests of similar samples by different operators was arranged in the summer of 1943. The samples of Ohio were some of the same lot reported in Table 3. The varieties Grundy and Adams grown in Michigan were carefully sampled to give comparable lots. The results of these tests given in Table 5 show no greater variability between the scores of the two operators for any one variety than between tests by the same operator and indicate that it is possible for different operators to obtain comparable scores on duplicate samples provided great care is used in treating and cracking the samples.

The differences in average score between the different varieties is consistent and apparently gives a correct indication of their relative merit. Grundy shows an average score of 93.7, Ohio 89.7 and Adams 69.2. The high score of Grundy is related to the large size of nut and high per cent first crack. The low score of Adams is related to small size of nut and low per cent first crack resulting from a large number of bound quarters. The kernels of this variety were plump, filling the cavity of the shell full and shattered on cracking.

In Table 6 are given the results of 54 tests of 38 selections or clones. In general it appears that the score is a fair indication of the worth of the sample. Low scores are related mostly to low per cent first crack and to the presence of empty nuts or shrivelled kernels in the sample. It is evident also that if a sample is too dry with many varieties a low score will result. Just what soaking treatment is most expedient is not too clear. Soaking 12 hours and drying 24 proved to be a satisfactory practice. The method followed by Mr. Stoke of soaking for 5 minutes and keeping the sample in a wet burlap sack for 24 hours is all right but is cumbersome if many samples are to be tested. Soaking one hour and holding 24 hours in a closed container like a coffee can give good results but percentage should be figured on dry weight and kernels should be air dried for 24 hours before weighing.

One weakness in the schedule is that it tends to give a small nut an advantage if the per cent kernel obtained in first crack is high. Thus a sample of the Mintle grown in Iowa which weighed but 13.6 grams per nut and total per cent kernel of 32 scored 101.1 points chiefly because the per cent first crack was 31.5. The same variety grown at Ithaca weighing 13.7 grams per nut but with 23.9 per cent first crack and 24.3 total scored 83.8. Possibly a penalty could be taken for nuts weighing less than 18 grams. On the other hand a large nut like the Grundy weighing about 23 grams would have a 10 point score advantage over Mintle and this may be enough for this character.

The six samples of Thomas grown on different trees in Ithaca, N. Y. in 1942 show great variation in score as has been the case in other years. Poor scores are related to shrunken kernels and such samples come from trees that are making poor growth because of poor soil conditions and competition with weeds. Also shriveled kernels are the result of defoliation by early frosts which may be very local and affect some trees and not others.

TABLE 6

Tests and Scores of Black Walnut Varieties from Various Sources.
25 nut samples unless otherwise indicated.
All scores figured on basis of 25 nuts.

• D—Dry
• S—Soaked
• No.—Hours dried or soaked

DISCUSSION

In the light of the data presented some conclusions can be drawn on the various questions raised at the beginning of this paper. It is evident that if approximately the same score is to be obtained by one operator on duplicate or replicate random samples, great care must be used in sampling. There is a tendency in taking samples to pick out the larger nuts or in some other way fail to take a good random sample. Selections submitted for contests are likely to be quite misleading as to the value of the variety and reflect in considerable part the contestant's skill in selection rather than the merit of the clone. The Freel walnut seems to be an example of this. At least as grown at Ithaca it is very disappointing.

It is evident that if comparable scores are to be obtained the samples receive the same treatment particularly as regards moisture content. Samples should be dried sufficiently to show the shrinkage of poorly developed kernels but in no case be allowed to dry to the point of checking the shells. Uniform soaking practice is a step in the right direction. A green or partially dried nut will test much higher than one properly cured as evidenced by Snyder, sample 6 and Spear, sample 7 in Table 1.

It seems probable that no schedule can be devised that will eliminate the necessity for skill on the part of the operator. To obtain satisfactory uniformity in scores, it is essential that the operator be skilled in the use of the cracking machine and use continuous care in applying the necessary pressure and in holding the nut in the anvils. Undercracking or overcracking, reversing the ends of the nut in the anvil or failure to hold the nut vertical may affect the score.

The presence of empty or poorly filled nuts in a lot of nuts from which samples are taken at random introduces greater variability in the samples than that found in lots with all nuts filled. This is true because the chances of getting an equal number of empty nuts in 25 nut samples are small and the presence of each empty nut decreases the per cent kernel and also the numbers of quarters possible. Variations due to empty nuts could be eliminated by greatly increasing the number of nuts in the sample but this is not practical for the purposes this schedule is intended to serve.

The question of whether or not it is possible for different operators to obtain equal scores on duplicate samples is not satisfactorily answered by the data in table 4. As the data stand the scores are far from equal. There is, however, a consistency in the scoring of each operator and it is quite probable that with more uniform treatment of nuts before cracking and more careful sampling better agreement would be achieved. This is borne out in the data given in table 5 in which the variation in scores between the two operators was no greater than that obtained by the same operator.

From a study of the data secured it appears that the causes of variation in the scores of duplicate or replicate samples are the result of (1) lack of care in making replicate random samples, (2) differences in treatment of samples before cracking, particularly as regards moisture content, (3) differences in the skill or care of the operator making the tests, (4) presence of empty nuts or shrivelled kernels in the sample which introduces variation not compensated for in a 25 nut sample and further complicates the matter because assigning penalties for shrivelled kernels involves personal judgment.

The first three of these can be minimized or eliminated by care and skill. The fourth item is not so easy but procedure can at least be standardized. Increasing the size of the sample is not practical if much testing is to be done.

All things considered it would seem that the scores indicate fairly well but not accurately the relative merit of the samples and thus can be relied upon to determine the relative merit of a variety or clone, the suitability of the variety for growing in a given locality and the variability of a variety grown in the same region but under different conditions. To determine the merit of a variety as compared to another both must be grown under the same conditions. The over-all value of a variety can only be determined from samples of well filled nuts. In any case the more samples tested the better.

The following suggestions are made as to procedure:

1. In taking a random sample no selection as to size, uniformity, or any other quality should be made. Suggested procedure would be to scoop up about 25 nuts in a berry basket or with the hands from the main supply and reduce the sample to 25 without conscious selection. What we in the Northern Nut Growers' Association want is a measure of the merit of the crop of the tree or variety in question and not the value of a highly selected sample.

2. It is not practical to bring samples to a uniform moisture content before cracking is done. The following precautions, however, may be followed: (a) Take care to see that nuts are reasonably well cleaned and free from fragments of husk. Scrubbing or beating the nuts together in a sack will usually remove most of the loose material. Of course the best practice is to wash the nuts immediately after shucking. (b) Cure samples until they are dry enough not to lose more weight preferably in an unheated room. This takes at least a month or 6 weeks. (c) Avoid storing the samples in a heated room where they will become so dry that the shells will check or crack. If this occurs the normal cracking fracture of the shell is destroyed and a satisfactory test cannot be made. (d) Nuts that have become so dry that the kernels shatter may be moistened by soaking about 2 hours in cold or lukewarm water then holding them in a moist condition for 18-24 hours, followed by drying for 10-12 hours before cracking. Nuts that are to be soaked should be weighed before soaking and the dry weight used in figuring percentages. The kernels of soaked nuts should be dried for 24 hours before weighing, preferably under the same conditions in which the samples were stored before weighing.

3. Care and skill on the part of the operator are of the greatest importance, particularly in the thoroughness of cracking. The most important variable in the score is the per cent kernel recovered at first cracking. The score is reduced by undercracking the nut so as to leave the quarters bound or by overcracking to the point of smashing the kernels. If the nuts have a long point so that the rims of the anvils do not contact the shoulders of the nut, poor cracking will result. At the present time a cracker with interchangeable anvils is not available. Using different sized iron pipe couplings in a vise may help solve the problem. Some varieties will crack better with a hammer than with a cracker of the Hershey type with standard anvils. In cracking a sample for test the operator should try to recover the most possible out of the first crack without using a pick or recracking.

4. The empty nut problem is probably the most difficult and is not satisfactorily solved by cracking nuts in excess of 25 until 26 filled nuts are secured. This necessitates weighing the sample after the nuts are cracked which is usually impracticable because of loss of parts of shells in cracking and because additional nuts are not available. Empty or shrivelled nuts in a sample are a serious defect which should count heavily against it. On the basis of experience it seems that a better method is to crack the random sample of 25 nuts and let the empty nuts and shrivelled kernels affect the score as reduced weight per nut, reduced per cent kernel and the penalty as well. Shrivelling that is obvious and which adversely affects the appearance of the kernels should be penalized. Possibly further experience will suggest a better way of handling this problem.

The proposed score of a sample is made up as follows:

1. The weight of a single nut in grams.

2. The per cent kernel of total weight of sample recovered after first crack x 2.

3. The total per cent kernel of total weight of sample divided by 2.

4. One tenth point for each whole quarter recovered.

5. Penalty of one score point for each empty nut in the sample.

6. Penalty of ½ point for every nut with shrivelled kernel.

The makeup of this score does not differ from that previously used except in the matter of procedure with empty nuts. It is felt that the items included are weighed in a realistic manner and that difficulties in scoring have been due to methods of handling the samples rather than in the scoring schedule itself. It does not seem likely that this schedule or any schedule will be valuable unless used by experienced operators who are willing to take the precautions indicated. Also it is apparent that wherever possible more than one sample of a lot to be scored should be tested and the average score used.

REFERENCES CITED

1. MacDaniels, L. H. Report of committee on varieties and judging standards. No. Nut Growers Assn. Proc. 28: 20-23. 1937.

2. MacDaniels, L. H. Is it possible to devise a satisfactory judging schedule for black walnuts? No. Nut Growers Assn. Proc. 30: 24-27. 1939.

3. Kline, L. V., and S. B. Chase. Compilation of data on nut weight and kernel percentage of black walnut selections. Am. Soc. Hort. Sci. Proc. 38: 166-174. 1941.

4. Kline, L. V. A method of evaluating the nuts of black walnut varieties. Am. Soc. Hort. Sci. Proc. 41: 136-144. 1942.

5. Lounsberry, C. C. Measurements of walnuts of United States. No. Nut Growers Assn. Proc. 31: 162-167. 1940.

6. Drake, N. F. Judging black walnuts. No. Nut Growers Assn. Proc. 22: 130-137. 1931.

7. Drake, N. F. Black walnut varieties. No. Nut Growers' Assn. Proc. 26: 66-71. 1935. Nut Growers Assn. Proc. 30: 81-83. 1939.

---