other hand delivering to the infringer a simple but infallible receipt for avoiding the patent.

Does the defendant infringe?

The point is made that infringement is avoided because the defendant introduces the alum into the intake-main and not directly into the filter. The court is of the opinion that so long as the alum is introduced simultaneously with the passage of the water to the filter and produces the same result in the filter-bed the precise locality of its introduction is immaterial. The principal controversy, however, arises over the function performed by the defendant's tanks, so that the issue may be narrowed to the single question, Does the defendant use settling. tanks! That it uses tanks is conceded; but are they settling-tanks? Does it use tanks "commonly employed" as settling-tanks at the date of the invention? Does it use tanks in which sedimentation takes place to any appreciable extent-tanks in which the work of purifying is carried on so as to relieve the filter-beds in any practical degree? In short, can it be said that the pure water produced by the Niagara plant is the joint product of the tanks and the filter? If such water be produced by a process of sedimentation in the tanks and a subsequent process of filtration, the Hyatt claim is not infringed. If, on the other hand, the purity of the water is due to the filter, precisely as if the intake pipe communicated directly with the filter, the claim is infringed. Running the water through the enlarged main does not avoid infringement, and this is so even though the water in its passage deposits a small and wholly inconsequential amount of sediment.

The foregoing is, it is thought, a fair statement of the issue as it relates to infringement.

The capacity of the defendant's tanks is said to be twenty-eight thousand gallons. The daily output of the plant is about three million five hundred thousand gallons, or about two thousand five hundred gallons per minute. This immense volume of water passes through the tanks daily and remains therein only about thirteen minutes. It is always moving. It is never at rest. Not only does it move longitudinally, but in almost every other direction as well. The water enters near the bottom of the tank at one end. It then flows over a baffle, through an archway, over another baffle, and is fiually pumped up through a large suction-pipe in the last compartment. That all this produces eddies, currents, cross-currents, vertical currents, and general turmoil in the tank is undisputed. In short, the passage of the water through the tanks resembles in a less degree its flow through the turbulent and rock-troubled channel from which it is originally taken.

The tanks are washed not more than once a week. At the time of Professor Main's visit, October 28, 1896, he says:

I could not find that anyone about the establishment knew when either of the tanks had been cleaned out. It might have been weeks or months for all the information I could obtain.

The filters, on the contrary, are washed twice every day.

If the proof stopped with these general facts, it would seem to the ordinary lay mind that these tanks, which hold but fourteen thousand three hundred and eighty gallons each and through which over three and a half million gallons flow each day in a confused and eddying mass, can hardly be the settling-basins of the prior art. But this opinion is concurred in by two men who stand at the head of their profession and upon whose judgment in such matters the court has a right to rely with confidence. Dr. Chandler says, in speaking of the defendant's tank:

It is evident, therefore, that this is not a settling-basin, or settling-tank in fact, and I am satisfied from my knowledge of the subject that it is impossible for a basin constructed as this one is constructed, and of the dimensions of this basin, taken in connection with the flow of water, to act as a settling-basin. In the first place, it is far too small for a filtering plant of the capacity of this one, even if the two basins were used alternately and the flow of water were stopped for a time in each basin successively; and, further, it is constructed in such a manner as to make it practically impossible for any sediment to accumulate in it, as it is used at Niagara.

President Morton, after examining the drawings of defendant's plant and Professor Main's description, says:

I am able to say with certainty that such a plant fully embodies the invention assigned to Hyatt by the above-quoted opinion of the court, and that the so-called "settling-basins" represented as connected therewith are in no sense settling-basins such as are referred to in the opinion of the court, and do not in any way modify the process of continuous filtration as defined in said opinion. In these so-called "settling-basins," the water is never in a state of rest, but, on the contrary, is in a rapid condition of motion, and not only so, but by special arrangement of cross walls or partitions, scouring-currents are established in the water as it moves through, well adapted to prevent sedimentation or settling of any solid matter which might exist in the water.

Persuasive as is this testimony, infringement does not rest upon opinion, but upon facts. The plant has been twice examined to determine this question-on October 28 by the complainant's experts and on December 7 by the defendant's experts. The examination on October 28 is the more important, first, because the tests were simpler and more determinative; second, because it was made in the presence of the defendant's expert, and third, because if infringement is established on the 28th of October it is no answer to say that there was no infringement on the 7th of December, when the conditions were wholly different both as to the turbidity of the water and the amount of alum used. If the Hyatt process were used on October 28, it must have been used for a number of days prior and subsequent thereto. This is sufficient to establish infringement.

It should be remembered that at the time of complainant's examination no one knew when the tanks had been last washed out. An immense volume of water had therefore passed through them. The court will take judicial notice of the fact that in such circumstances impure water will deposit some sediment. The examination of Professor Main and Mr. Kendrick demonstrates conclusively that the defendant does not use settling-tanks. Substantially the entire surface of the

basin was perfectly clean. In one of the compartments were two small patches of mud half an inch deep where it lay thickest. The entire amount of mud in the whole basin was estimated at less than half a bucketful. The water when examined near the bottom with an electric light showed no evidence of turbidity due to sedimentation. When the tank was drained, the last water that ran out was clear.

An examination of the samples of water taken before its entrance to the tanks and after it had passed through them showed practically no difference in turbidity. In short, the facts found were absolutely incompatible with the theory of settling-tanks, considering the size of the tanks, the volume of water, and the small quantity of alum used.

But this was not all. An examination of the filters showed that substantially the entire work of clarifying and purifying the water was done in the filter-bed, precisely as in the Hyatt process. If there had been any substantial sedimentation in the tanks, the result would have been registered in the filter-bed.

Assuming Professor Main's observations to be correct, the court has no doubt that the defendant is practicing a continuous process of filtration by the use of small quantities of coagulant-too small to produce substantial sedimentation in settling-basins-the impurities being removed by passing the water through a filter-bed of sand-in other words, the Hyatt process. The correctness of these observations is hardly disputed so far as their principal features are concerned.

The chief criticism of Professor Main's method is that he used the suction-pump to drain the tank; but as over two feet of water remained when the pump stopped operating, and as this was the water examined and found to be clear, it is thought that there is little force in the criticism. Again, it is said that the statement that the tank was perfectly clean is discredited by the fact that on coming up from the tank the experimenters washed their hands. This certainly proves that they washed their hands, but it proves little else.

The facts found by Professor Main and Mr. Kendrick are hardly disputed at all. The contradictions are in the most general and guarded terms. One of the witnesses saw mud and slime in the channel in the middle of the tank. How much he does not say, except that it was a "thin layer." Another speaks of "a substantial deposit of slime and mud." Another says that there were 66 many times 'two quarts' or 'a half bucketful' of mud" in the tank, and testifies further:

I have examined these settling-basins on other occasions and have invariably found a substantial deposit of mud and slime; the amount always being in proportion to the turbidity of the water since the last cleansing of the settling-tanks.

Two other witnesses testify to having seen, when the tanks were being cleaned, "a noticeable deposit of sediment and mud," and a third has seen "thick mud and slime deposited in the settling tanks." This is uncertain and vague and wholly inadequate to disprove the posi tive assertions of the complainant's witnesses. The statement that a muddy stream will deposit more mud than a clear one is thought to

be in exact accordance with the truth, and accounts for some of the different results found on December 7.

For the reasons heretofore stated it is unnecessary to discuss the examination on the 7th of December, when no one representing the complainant was present. The difference in the quantity of mud may well be accounted for by the increased turbidity of the water and the length of time the tank had been in use without cleaning. The experi ments of the 7th, though interesting, do not seem to the court to destroy the verity of the conclusions reached by the complainant's experts. The court is therefore of the opinion that infringement has been established.

The motion is also opposed upon the ground that new evidence has been adduced tending to show anticipation. Several of the affidavits relate to alleged prior use by Benjamin T. Loomis and Livingston H. Gardner, at Baltimore and New Orleans, respectively. Substantially the same facts were presented to the court on a motion to open the Schwarzwalder decree. (62 Fed. Rep., 582.)

Gardner's contribution to the art was summarily dismissed with the following observation:

The New Orleans affidavits are, in my opinion, unimportant. The information which they contain is unsubstantial in its character.

As to the Loomis defense, it is true that the motion was denied because the defendants had been guilty of negligence in not presenting it sooner; but it is thought that had the learned judge who decided the cause at circuit been impressed with the idea that Loomis's testimony would have invalidated the patent so careful a jurist would have found some way to have the evidence brought before him.

The affidavit of Mr. Loomis is largely devoted to excuses for not using the invention after it was made and for not having it patented. Neither excuse seems entirely satisfactory. He sold many filters in Baltimore, but did not attach his alum-feeding device, because

the water was so pure and free from disease germs as to require no special care in filtration.

This would seem a plausible reason for selling no filters at all; but, as the citizens evidently thought their pure water needed filtering, it would seem that Baltimore was an ideal market for an improved filter. After December, 1882 (he does not say how long after,) he had an order for a filter to be used at a place a distance from Baltimore, to which he attached the alum-feeding device; but he omits to state where the place was or who gave the order. He had obtained a patent for a filter in September, 1880; but he did not patent his invention of 1882, because he thought it covered only the particular form of apparatus, and this he did not consider worth patenting.

It seems incredible that if Loomis had actually discovered the Hyatt process his achievement should have reached a termination so lamentable.

15377-21

The single filter made by Peterson was, on his own showing, a failure. There is no evidence that it is in existence, and the description is too indefinite and uncertain to be considered.

The other anticipatory evidence has been examined; but it is thought that nothing material has been added to the record in the Schwarzwalder case. It may be said generally of all this proof that it is shadowy and uncertain and fails to meet the rule so often reiterated by the courts that prior use must be established beyond a reasonable doubt. This motion was submitted after unusually able oral arguments, but without briefs or any assistance from complainant's experts so far as this branch of the controversy is concerned. As the affidavits relating to prior use were first presented at the argument, this was inevitable. The court has examined this testimony with all the care possible in the circumstances and entertains the hope that nothing important has been overlooked. Although some parts of the testimony have not been discussed, all have been considered.

The complainant has established its patent after years of fierce and expensive litigation. The patent has but four years more of life. If relief be withheld now, the complainant is practically remediless. Its business will be destroyed long before the second weary journey through the courts is terminated.

The equities are with the complainant. So are all the presumptions. If either party must suffer pending the final decree, it should be the defendant and not the complainant.

The motion is granted.

NOTE. If the injunction be issued immediately, it may cause great inconvenience to the citizens of Niagara Falls and endanger their health as well. If the appeal be speedily prosecuted, it would seem that a decision may be reached within a few months. In the meantime it is suggested that the parties agree upon a modus vivendi which will preserve the rights of all.

[Court of Appeals of the District of Columbia.]

ROSEBRUGH ~. Holman.

Decided December 17, 1896.

78 O. G., 1258.

ROSEBRUGH-FAILURE TO FILE APPEAL-APPEAL DISMISSED-COSTS.

When it appears that an appellant has failed to have his cause filed and docketed. his appeal may, on motion, be docketed and dismissed, with costs.

APPEAL from the Commissioner of Patents.

Messrs. Foster & Freeman for the appellant.

Mr. Reuben L. Roberts for the appellee.