dissenting, with whom MARTIN, J., joins.
The point which forms the principal reason for my disagreement with the majority opinion is the inordinate weight ■given appellant’s affidavit.
Appellant’s specification, as illustrated by the quotations set out by the majority, points out that use of 98.5% to 100% p- or m- xylene results in a decrease in by-product formation as well as yields which are high and “entirely unexpected” when compared to the ordinarily expected theoretical yield and space-time yield. I consider those statements in the specification to be well tempered by appellant’s concession in his affidavit, also quoted by the majority, that:
By means of the present improvement, i. e., by the use of 98.5 to 100% xylenes, not only are high yields in the end products and less by-products obtained, as is to be expected, but the present improvement also constitutes a distinct advantage in that a high space-time yield1 is attained. * * * [Emphasis supplied] .
The examiner noted:
* * * The result of using purer materials is precisely that expected, i. e. an increase in the yield with less side-reactions. * * *
and appellant, in his brief before the board, stated
* * * One expects a higher yield when using a purer starting material.
In view of the results expected when one uses purer starting material, it seems to me adequate reason has been provided to do what appellant has done. Weighing all the evidence, I do not find reversible error in the board’s conclusion that the subject matter as a whole is obvious to one of ordinary skill. I would affirm.
. Neither appellant nor the Patent Office has furnished a definition of space-time yield from which we can evaluate the merits of the contentions with respect thereto. Perry’s Chemical Engineering Handbook, 3rd Edition, states at page 329:
* * * Space velocity is defined to be the volume of gas * * * or liquid passing through a given volume of catalyst space v0 in unit time divided by the latter, i. e. space velocity = V/vc. * * * The yield of desired product in unit time per unit volume of catalyst per passage is the space-time yield. This is the product of the fractional conversion F by the space velocity, space-time yield = F(V/v0).
Increasing the purity of starting material would necessarily appear to increase the value of at least F in the above equation, thereby increasing space-time yield.