Our conclusions differ in a few points from those of our first work on this subject. We are able to sum up our observations in this series of experiments as follows: 1. The ophthalmo-tuberculin test is of limited value in the diagnosis of tuberculosis in cattle. In some cases the reaction is very slight (hyperaemia). In others more pronounced congestion with profuse exudates are noted. Accuracy of observation is important. We are inclined to rely primarily on the results of the first instillation of tuberculin. Second instillations in a few instances elicit reaction in non-tubercular animals. 2. In the majority of animals tested the reaction increased in its intensity with each subsequent instillation of tuberculin. This fact indicates the development of a local hypersusceptibility or anaphylaxis associated with a partial immunity; von Pirquet calls this condition "allergie" (9). 3. It is possible in some cases to create a condition of "allergie" in healthy cattle, when spaced instillations of tuberculin are made. It is evident, therefore, that the result of the first instillation of tuberculin should be made the only basis of diagnosis. Rosenau and Anderson (II) have recently called attention to this point in regard to the human subject. 4. When repeated instillations of tuberculin are made on the conjunctiva at short intervals (twenty-four hours, etc.) a local immunity results (No. 3D et al.). If the instillations are separated two weeks or more anaphylaxis results. 5. We, therefore, hold that if tuberculin (0.1 cubic centimeter) is carefully instilled into the conjunctival sac and if careful comparison of the instilled eye with the opposite eye shows that a reaction of varying intensity results in from ten to twelve hours after the first instillation, a tubercular lesion is present. 6. In our first report (7) we were inclined to believe that subcutaneous tubercular injection given previous to the ocular test would slightly inhibit it. We have since become convinced that this is true only to a limited extent, and that in some cases the ophthalmo-reaction is exaggerated by a subcutaneous injection of tuberculin. 7. The primary ophthalmo-tuberculin reaction is in direct proportion to the extent of the tubercular processes in the body. The more extensive the tubercular processes, the more anaphylactic the animal is. This is in direct variance with the condition in the usual subcutaneous tuberculin test. (See Necropsy Report.) 8. We are inclined to believe that the ophthalmo-tuberculin test will reveal tuberculosis at as early a state as the usual subcutaneous test. 9. The ophthalmo-reaction is of no value in determining whether vaccinated cattle are actively tubercular or not, or in demonstrating any hypersusceptibility in the offspring of tubercular cattle. 10. The cutaneous test from our brief series of experiments does not seem to be as accurate as the ophthalmic test. This conclusion has been reached by several investigators.