241 F2d 944 Conner v. G Joris

241 F.2d 944

Joshua C. CONNER, Appellant,
v.
George G. JORIS, Appellee.

No. 6225.

United States Court of Customs and Patent Appeals.

February 21, 1957.

Kenyon & Kenyon, New York City (Clinton F. Miller, Wilmington, Del., Solon B. Kemon, Theodore S. Kenyon, and Malvin R. Mandelbaum, New York City, of counsel), for appellant.

Gordon A. Wilkins, New York City (George B. Campbell and Robert A. Harman, New York City, of counsel), for appellee.

Before JOHNSON, Chief Judge, and O'CONNELL, WORLEY, RICH, and JACKSON retired, Associate Judges.

JOHNSON, Chief Judge.

1

This is an appeal from the decision of the Board of Patent Interferences of the United States Patent Office awarding priority of invention, as claimed in the single count hereinafter set forth, to the senior party, George C. Joris (hereinafter referred to as Joris). The single count in issue reads as follows:

2

"In a process for oxidizing cumene to cumene hydroperoxide in liquid phase with elemental oxygen as oxidizing agent, the improvement which comprises maintaining solid sodium bicarbonate in contact with said cumene, at reaction temperature in the range between about 60°C and about 90°C."

3

The present interference was originally declared between Joris patent No. 2,577,768, filed July 19, 1949, and issued December 11, 1951, and a Lorand and Reese application No. 279,241, filed March 28, 1952, which application was assigned to Hercules Powder Co. This application was a continuation-in-part of application No. 257,671, filed November 21, 1951, which, in turn, was a continuation-in-part of parent application No. 31,396, filed June 5, 1948.

4

Counsel for Lorand and Reese moved to substitute application No. 317,589, filed by Joshua C. Conner, Jr. (hereinafter referred to as Conner) on October 29, 1952, for that of Lorand and Reese, alleging that Conner had been found to be the true inventor as against Lorand and Reese. This motion was granted.

5

Testimony was taken on behalf of Conner, but Joris took none. Both parties filed briefs and made oral arguments before the board.

6

The party Joris relies for priority of invention on the filing date of the application for his patent.

7

The record indicates, and this point was not contradicted, that Conner conceived the invention here involved in February, 1948. He relies on an actual reduction to practice, based upon a series of tests conducted in May, 1948.

8

There is no question as to the fact that the count is adequately supported by the disclosures of each of the parties in suit.

9

Joris originally contended, before the board below, that Conner is estopped from claiming priority of invention by reason of his delay in filing (more than four years), but this contention was disposed of by the board and is not pressed here by Joris.

10

The board, having found that Conner had not sustained the necessary burden of proof of actual reduction to practice, concluded that priority of invention must be awarded to Joris.

11

The sole question here is whether the tests conducted by Conner in May, 1948, constituted an actual reduction to practice. If so, priority of invention must be awarded to Conner.

12

Since Conner's filing date is subsequent to the issue date of the Joris patent, Conner has the burden of proving priority of invention beyond a reasonable doubt. Walker v. Altorfer, 111 F.2d 164, 27 C.C.P.A., Patents, 1130.

13

It appears from the record that Conner, and others who testified on his behalf, were, at the time of the alleged reduction to practice, in the employ of the Hercules Powder Co., which is the real party in interest. Conner's testimony clearly establishes that his work directed towards the process recited by the count in issue had a direct bearing on commercial operations then in progress at the Hercules plant in Brunswick, Georgia. He testified that it was his intended purpose to find a method of oxidizing cumene to cumene hydroperoxide in high yields. In light of the foregoing testimony, it is as well evident that his purpose was to increase yields at the plant.

14

Though the cumene then in use at the plant was impure and was subjected to no special pretreatment prior to its introduction into the reaction mixture from which the cumene hydroperoxide product was formed, all of the cumene used by Conner in his experiments was specially pretreated.

15

The testimony as to the effect of the pretreatment on the subsequent cumene oxidation process was conflicting and indicated that said pretreatment might affect both the yield and the reaction rate of the process.

16

On the strength of the foregoing, the board concluded that Conner had not sustained his burden of proving practical utility pursuant to the intended purpose of the process recited in the here involved count, citing as its authority Balogh v. Crot, 176 F.2d 923, 37 C.C.P.A., Patents, 707; Powell v. Poupitch, 167 F.2d 514, 35 C.C.P.A., Patents, 1080; Burns v. Curtis, 172 F. 2d 588, 36 C.C.P.A., Patents, 860; Saklatwalla v. Marburg, 172 F.2d 227, 36 C.C.P.A., Patents, 791; Landon v. Ginzton, 214 F.2d 160, 41 C.C.P.A., Patents, 950.

17

Quoting directly from the Balogh case,1 [176 F.2d 926] the board concluded that since the cumene used by Conner as a starting material in his experiments was different from that used in the operations of the Brunswick plant, and since Conner had not proved that the elaborate pretreatment had no effect on the oxidation rate or yield of the cumene, his tests did not simulate the actual service conditions in the process which he was seeking to improve upon and were therefore not sufficient to constitute an actual reduction to practice.

18

We are of the opinion that the board has erred in its application of the test of utility.

19

It is elementary that, excepting plants and designs, an invention is not reduced to practice until its practicability or utility is demonstrated, Rivise & Caesar, Interference Law and Practice, Vol. 1 (1940), § 138, and that this utility is that pursuant to its intended purpose, Landon case, supra.

20

In those cases where actual tests are required to demonstrate the practicability or utility of the invention, the tests must simulate actual service conditions with sufficient clearness to render it reasonably certain that the subject matter will perform its intended function in actual service. Chittick v. Lyons, 104 F.2d 818, 26 C.C.P.A., Patents, 1382; Powell v. Poupitch, supra. It is immaterial that the tests employed were on a laboratory scale, provided the foregoing requirements are met. Chittick case, supra.

21

In the instant case, the process of oxidizing cumene to cumene hydroperoxide by means of air or molecular oxygen was known, as is admitted by Conner. There is no question but that the inventive feature of the count resides in the use of solid sodium bicarbonate in the oxidizing process. The purpose of the sodium bicarbonate may be gleaned from the following portion of Conner's specification:

22

"* * * none of the previously known processes have been successful in effecting substantial yields of α, α- dimethylbenzyl hydroperoxide [another name for cumene hydroperoxide] alone. Under the conditions practiced in prior procedures the hydroperoxide has not been obtained in good yield and to the exclusion of other reaction products. The oxidation instead has led to mixtures containing substantial amounts of acetophenone and α, α- dimethylbenzyl alcohol.

23

"Now in accordance with this invention it has been found that α, α- dimethylbenzyl hydroperoxide may be prepared to the practical exclusion of secondary reaction products * * *."

24

And, in another portion of Conner's specification it is said:

25

"The process of this invention produces high yields of α, α-dimethylbenzyl hydroperoxide while at the same time minimizing the formation of other reaction products * * *."

26

Neither in Conner's specification nor in the count do we find any statement which adds, insofar as intended purpose is concerned, to the aforequoted purposes of the Conner process. Nothing is said, either in the specification or count, about the use of either "pure" or "impure" cumene. The word "cumene," wherever used by Conner, is absolute and unqualified.

27

In going beyond both the count and the specification to glean Conner's intended purpose the board has gone far beyond any position supported by the cases cited or any that we have been able to find.

28

As was stated in the Landon case, supra [214 F.2d 163]:

29

"`The patent application may reveal a number of purpose for the same invention. The inventor need prove only one practical use; only one useful result or effect. * * *'"

30

In the instant case, the "actual service conditions" necessary to constitute the test otherwise acceptable to prove actual reduction to practice do not refer to the use of the same purity cumene as was used in the commercial operations at Hercules but to the operating conditions, such as temperature, etc., which would be employed if the pretreated cumene of Conner were used on a commercial scale. We have no doubt that these conditions were, in fact, employed in the tests conducted by Conner and his associates.

31

It remains to be decided whether or not Conner's tests were sufficient to demonstrate the utility of the here involved process, viz — whether the tests indicated either an increased yield of cumene hydroperoxide or a minimizing of the formation of other reaction products.

32

The record indicates that the essential experiments relied upon by Conner to establish actual reduction to practice (the May 3-6, 1948, run) were performed by one Hicks, a research chemist employed by Hercules, under the direct supervision of Conner. Extensive notebook entries of the observations and results of these experiments were made at the time of the run, and both Conner's and Hicks' names were signed on each page of the notebook on which entries were made. (Conner Exhibits Nos. 4A, 4B, 4C.)

33

At periodic intervals during the run, samples were removed by Hicks and sent to the Analytical Division of the Hercules Experiment Station for analysis. The analysis included both a refractive index determination of the sample, from which the weight per cent of oxidation products was determined, and a chemical test, from which the weight per cent of cumene hydroperoxide was determined.

34

The results of these analyses were communicated to Hicks by telephone, which results were recorded in the laboratory notebook. Formal reports prepared by the Analytical Division subsequently were sent to Hicks. (Conner Exhibits Nos. 7A and 7B.) Though Hicks testified to the effect that he had made an independent chemical analysis on samples in the May 3-6 run to determine the per cent of cumene hydroperoxide in said sample, the record indicates that it was the practice of both Conner and Hicks to rely on the reports from the Analytical Division for the data relating to the results of the experiment.

35

The record as well includes evidence of a test run performed by Hicks under the direct supervision of Conner in April, 1948, which run was similar in every respect to that of the May 3-6, 1948, run except for the fact that no sodium bicarbonate additive was present in the earlier run. (Joris Cross Exhibits Nos. 2B and 2C.)

36

Also introduced into evidence was a formal report signed by Conner and sent by him to the management at Brunswick (Conner Exhibit No. 5) and a formal report signed by both Conner and Hicks and circulated to several of Conner's superiors at Hercules (Conner Exhibits Nos. 6A and 6B.)

37

The board was of the opinion that Exhibits Nos. 5, 6A and 6B did not serve to prove an actual reduction to practice for several reasons, included in which was a failure of the documents to prove conviction of success, incompetency of the evidence in that they were self-serving declarations of the inventor and a lack of authentication thereof by testimony independent of the inventor.

38

We find it unnecessary to rule upon the correctness of the latter two of the above three statements of the board (the first statement will be hereinafter treated) as we are of the opinion that these documents are inadequate to prove Conner's actual reduction to practice for more basic and compelling reasons.

39

We will assume arguendo that the latter two evidentiary holdings are improper and that the documents may be given their due weight as admissible evidence.

40

At this point it is appropriate to note that counsel for Conner raises several questions as to the propriety of the board's statement to the effect that:

41

"Conner relies upon Exhibits 5, 6A and 6B to support his contention that he obtained a yield advantage by using sodium bicarbonate in the May, 1948, experiment. * * *"

42

It is strenuously asserted that Conner does not rely on these exhibits to support such contention but merely to show the form in which he reported the yield advantage to his superiors in Hercules, the data contained therein having been derived from data previously introduced into evidence.

43

We are of the opinion that it is immaterial what reliance was placed upon these documents by Conner since, as will hereinafter be demonstrated, these documents, taken either alone or in combination with the basic documents introduced as evidence, fail to meet Conner's burden of proof of actual reduction to practice.

44

It is urged by counsel for Conner that Exhibit No. 52 indicates the yield advantage obtained by Conner with the use of sodium bicarbonate. He relies heavily on the comparison in results of the data listed at 2% by-product formation to show that with sodium bicarbonate, the yield of oxidation products was 42.5% whereas without sodium bicarbonate, the corresponding yield was only 37.5%. He furthermore asserts that the 2% by-product formation represents only about 4.7% of the oxidation products (2/42.5 × 100) with sodium bicarbonate and 5.3% without (2/37.5 × 100).

45

As to the yield data, the 42.5 and 37.5% figures are misleading. A comparison of the results shown in the laboratory reports (Joris cross Exhibits Nos. 2B and 2C and Conner Exhibits Nos. 4A, 4B and 4C) representing the April and May runs,3 respectively, indicates that the yield of cumene hydroperoxide at comparative intervals in the April and May runs was greater in the former (without sodium bicarbonate) than in the latter (with sodium bicarbonate) at every period except that at 21 hours. No sample was taken of the April run at 70.5 hours and therefore the comparable reading of the May run sheds no light on the subject.

46

As to the percentage of by-products in the total oxidation products, it is significant that Conner is silent as to the data in Exhibit No. 5 at the higher reacted levels (the 5th, 6th, and 7th columns). It will be noted that at 50% reacted in the run with sodium bicarbonate the by-products represent approximately 9% of the oxidation products (4.5/50 × 100) whereas at 47.9% reacted, in the run without sodium bicarbonate, the by-products represent only 8% of the oxidation products (3.8/47.9 × 100).

47

When the data of the 2% by-products columns is thus regarded in the light of the higher level columns, it is seen that little strength can be accorded Exhibit No. 5 on this basis.

48

Thus, it is evident that the evidence incorporated in Conner Exhibits Nos. 4A, 4B, 4C and 5 is totally inadequate to demonstrate that Conner, in his tests of May 3-6, 1948, achieved the results he here asserts. On the contrary, they appear indicative of the converse of this, viz — that Conner was dissatisfied with this test and considered it to be nothing more than an unsuccessful experiment.

49

Consideration of Conner's Exhibits Nos. 6A and 6B does not alter our conclusion. Exhibit No. 6A, when construed most favorably to Conner, indicates only that the oxidation rates with and without the use of sodium bicarbonate were approximately equal. Exhibit No. 6B is merely a reiteration of the basic data found in Exhibits Nos. 4A, 4B and 4C. We find nothing therein indicating that either an increased yield or minimization of by-product formation was attained by means of the bicarbonate additive.

50

Conner testified to the effect that despite the fact that the oxidation rates with and without the bicarbonate were roughly equal, it was nevertheless advantageous to use sodium bicarbonate because thereby the formation of acidic by-products was avoided and also because a reproducible oxidation rate was obtained. It is difficult to see on what basis Conner concluded, by virtue of the May 3-6 run, that reproducible oxidation rates were obtainable where sodium bicarbonate was used. Except for an unsuccessful test run in February, 1948 (on which Conner relies for conception, supra), the May 3-6 run was the only run conducted in the presence of sodium bicarbonate. Surely, one successful test run (were the May 3-6 test run, in fact, successful) is not sufficient to establish reproducibility of results. Indeed, Conner's conclusion as to reproducibility seems to have been gleaned from the Joris patent, for no mention of such advantage is found either in the Conner specification or in any of the documents introduced into evidence by Conner. The board correctly concluded that such recourse to the opposing party's disclosure is insufficient to establish the independent fact of a successful actual reduction to practice, which must be proved affirmatively by the party asserting such fact. Brooker v. Riester, 161 F.2d 745, 34 C.C.P.A., Patents, 1088.

51

That no evidence was introduced which shows a minimization of by-products has heretofore been discussed.

52

Conner further argues that a subsequent test run conducted by one Filar, an employee of Hercules, in November 1953, more than 5 years after Conner's May 3-6 test run, was evidence tending to justify his (Conner's) earlier conclusions, and cites Guinot v. Hull, 204 F. 2d 281, 40 C.C.P.A., Patents, 982, as his authority. It is obvious that the Guinot case has no applicability here. For we are convinced that Conner had no positive "conclusion" in May of 1948 which was capable of being justified by any subsequent tests. If any conclusion might have been reached by Conner at that time, it was that his tests were not satisfactory to demonstrate the utility of his invention for its intended purpose.

53

Confirmatory of our conclusion that the party Conner has not adequately sustained his burden of proof are the facts relating to the attempts to secure patent protection on his alleged invention. It will be noted that a 4½ year hiatus separated Conner's May 3-6 test run and the filing of the application by him, which application contained the count here in controversy. Conner's only excuse for his failure to seek patent protection is that "I felt that we were covered in a general way for this type of oxidation system patent-wise." Conner was obviously referring to the Lorand and Reese applications, referred to supra. It is true that the Lorand and Reese application No. 31,396, filed June 5, 1948, claimed alkaline stabilizing agents broadly and disclosed several such agents. No mention therein, however, was made of the use of sodium bicarbonate. Nor did Conner make any mention of the possible use of sodium bicarbonate additive in his comments on the proposed patent application of Lorand and Reese, submitted to Conner for his remarks by the Hercules Patent Division (Conner Exhibit No. 13). It was not until 1951, when Lorand and Reese filed a continuation-in-part (application No. 257,671, filed November 21, 1951) of the patent application, that sodium bicarbonate was specifically disclosed, but not yet claimed. Finally, in application No. 279,241, filed March 28, 1952, which was filed by Lorand and Reese as a continuation-in-part of the preceding application, the use of sodium bicarbonate was claimed. As heretofore mentioned, Conner's instant application was substituted for the latter Lorand and Reese application.

54

We are aware of the fact that once an inventor has completely reduced his invention to practice, he is not required in an interference to show subsequent diligence in applying to the Patent Office for patent protection. Bowers v. Valley and Ernst, 149 F.2d 284, 32 C.C. P.A., Patents, 1039. However, it is also well established that when there is doubt as to whether there has been an actual reduction to practice, the inventor's subsequent conduct may disclose that, instead of a reduction to practice, the acts relied on show that what was done amounted only to an abandoned experiment. Bowers case, supra.

55

We are of the opinion that Conner's conduct subsequent to the May 3-6 run is further indicative of his consideration of said run as an abandoned experiment rather than as a successful reduction to practice.

56

It is also worthy of note that, after Conner's May 3-6 run and until the Filar test, supra, in November, 1953, no further tests were made using sodium bicarbonate. It is stated in the brief for Conner that:

57

"* * * no use was made of the process of the count because the commercial operations at Brunswick were conducted at a higher temperature. * * *"

58

The higher temperatures used at Brunswick, it is explained, were employed with a design to increase the productivity of cumene hydroperoxide (rate of yield) as distinguished from the yield thereof.

59

Conner testified, in answer to a question concerning the use at Brunswick of the higher temperatures, that:

60

"* * * The demand was so great that we were pushing the production as high as we could with existing equipment."

61

It is indeed strange, in view of the great demand at Brunswick for cumene hydroperoxide, that Conner's process was not employed in lieu of the existing process employing higher temperatures. While this, in itself, might not be conclusive of the issue, it tends further to establish the fact that Conner's experiment was not successful and was, in fact, abandoned by him.

62

Having reviewed the record thoroughly, we are constrained to conclude that the party Conner has failed to establish beyond a reasonable doubt that his tests were sufficient to demonstrate the utility of his process for its intended purpose. The evidence certainly does not indicate that he had conviction of success. See Saklatwalla v. Marburg, supra, 172 F.2d at page 231, 36 C.C.P.A., Patents, at page 799 and Muskat v. Schmelkes, 140 F.2d 984, 31 C.C.P.A., Patents, 837, for a discussion of the requirement of "conviction of success."

63

In our consideration of the issues involved in this case, we have not been unmindful of the extensive discussion made by both counsel for appellant and for appellee concerning the applicability of Reiners v. Mehltretter, 236 F.2d 418, 421, 43 C.C.P.A., Patents, 1019. That case was an interference involving a count drawn to a process for the production of 1,2-isopropylidene glucuronic acid from 1,2-isopropylidene glucose by means of an oxidizing agent. This court there stated:

64

"The experiment here under consideration was not of a pioneer nature. Both the starting products and the end products were well known compounds and the steps carried out were of a type which had been previously performed on generally similar materials. The reactions to be obtained could, therefore, be predicted with a reasonable assurance of accuracy and under such circumstances it is not necessary that the proof of identity of the products be as exhaustive as if entirely new substances or procedures were involved. Guinot v. Hull, 204 F.2d 281, 40 C.C.P.A., Patents, 982."

65

Upon the authority of this case, appellant contends that since processes identical to the instant process (except for the fact that other alkaline materials were used in lieu of sodium bicarbonate) were known and used prior to his alleged actual reduction to practice in May, 1948, the degree of proof necessary to prove such actual reduction to practice is diminished and that he has satisfied the reduced burden.

66

While we are in accord with what was set forth in the Reiners case, we cannot agree that appellant's case stands in better light as a result of what was there said.

67

In the first place, the question which elicited the foregoing statement from this court in the Reiners case was whether or not any 1,2-isopropylidene glucuronic acid had been produced. The count was not limited to any degree of purity, etc., of the acid, but only to its production in general. We did not there dispense with the necessity for tests which would establish the utility of the process for its intended purpose but, rather, held that the tests which were performed, in the light of the similarity of the involved process to prior processes, were sufficient to establish the production of the acid in question.

68

Here, however, the count specifically sets forth the improvement over the prior art process, which improvement consists in the use of sodium bicarbonate additive. Indeed, appellant does not question the propriety of considering the adequacy of his tests to establish that the inventive means specified in the count (viz. — the use of sodium bicarbonate additive) was successful for its intended purpose: the increase in yield of cumene hydroperoxide and/or minimization of by-product formation. The prior use of other alkaline stabilizing agents does not aid us in our determination of his success in his May, 1948, endeavor, nor does it lighten his burden. This is conclusively shown by the following excerpt from the record, at which point Conner was answering the questions of counsel for appellee:

69

"XQ344. Am I right that in your testimony you testified `We decided a buffer type sale [sic] was needed, and this led us to sodium bicarbonate'? A. Yes.

70

"XQ345. What was meant by a `buffer type salt'? A. In that instance it was meant a salt which had the power of absorbing an acid without undergoing a great pH change.

71

"XQ346. Again I suppose that means or am I right that you are thinking of its having that power under the reaction conditions? A. Yes.

72

"XQ347. Then am I right that you would have to actually test the sale [sic] in order to find out whether it had that power or not? A. It should be tested under the reaction conditions, I believe.

73

"XQ348. If a salt was named and you had never tested it, could you say whether it had this power or not? A. I could not precisely predict how it would function in this anhydrous oxidation system, since it is a non-ionic medium, but I could perhaps generalize as to whether I though it would give a high yield of hydroperoxide." (Emphasis added.)

74

It is not seen how a comparison with other processes of a similar nature, wherein other alkaline stabilizing agents were used in lieu of sodium bicarbonate, can aid appellant. No other conclusion can be reached in view of appellant's own admission that he could only "generalize" as to whether the use of sodium bicarbonate "would give a high yield of hydroperoxide." Certainly, the proof adduced in the Reiners case, supra, was more compelling than that offered by appellant. It is, as well, not an insignificant fact that in the Reiners case the burden of proof on the junior party was only that he establish his case by a preponderance of evidence, whereas here, appellant must so do beyond a reasonable doubt.

75

We conclude, therefore, that appellant has failed to establish an actual reduction to practice in May, 1948. No further questions having been raised on this appeal, priority must be granted to the senior party, Joris, on the basis of his constructive reduction to practice on July 19, 1949.

76

The decision of the Board of Patent Interferences is hereby affirmed.

77

Affirmed.

78

JACKSON, J., Retired, recalled to participate in place of COLE, J.

Notes:

1

"There is no limitation in any of the counts of intended use. * * * However, proof of actual reduction to practice, which demonstrates successful operation under the actual conditions in which it was designed to work, is necessary. * * *"

2

"Exhibit No. 5 represents a formal report prepared by Conner and sent to Mr. A. H. Reu, manager of the Brunswick plant, on July 22, 1948. Contained therein is a chart illustrating the comparative results of the April and May, 1948, experiments conducted by Conner for the production of cumene hydroperoxide, the former without and the latter with the addition of sodium bicarbonate. The data in this report was graphically obtained by Conner from the basic data recited in Joris Cross Exhibits Nos. 2B and 2C and Conner Exhibits Nos. 4A, 4B and 4C. The essential portion of this report reads as follows:

  "II. * * *

    "A. Effect of Acid Acceptors

  "* * *    Acid
          Acceptor     % Reacted     % By-Products  Oxidation    % Reacted     % By-Products  Oxidation
            (1%)     ("Conversion")    ("Spread")     Rate     ("Conversion")   ("Spread")       Rate    Recovery
  "* * *   NaHCO3         42.5              2         0.78            50            4.5          0.78      84.2
  "* * *    None          37.5              2         0.80          47.9            3.8          0.78      91.1

"* * * * * * *

"The 2nd, 3rd and 4th columns reprepresent the per cent reacted when the by-product formation obtained a maximum of 2% (the `% By-Products Spread' represents the difference between the % reacted (`Conversion') and the % cumene hydroperoxide in the sample) while the 5th, 6th and 7th columns represent the by-product formation at higher reacted levels."

3

"A comparison of the results recorded in these reports is shown as follows:

  "April 19-21, 1948                                   May 3-6, 1948
  "(Without NaHCO3)                                    (With NaHCO3)
  -------------------------------------------------------------------------
  "HRS    % Oxidation     % Cumene       HRS    % Oxidation     % Cumene
            Products   Hydro-Peroxide             Products   Hydro-Peroxide
             (Net)          (Net)                  (Net)          (Net)
  -------------------------------------------------------------------------
  " 5        3.17           3.1           5         2.13           2.0
  "21       14.1            5.2          21.8      13.1           14.3
  "29       20.9           21.2          29.16     20.2           20.65
  "45       37             34.6          45.75     31.9           34.4
  "56.25    44.9           41.1          53.8      42             39.1
  "70.5    No data                       70.5      52.8           45.5"