of the evidence furnished. Even the very dilute solution which I have been able to make of the material scraped from these spots gives characteristic absorption bands between the D and E lines, as they are called. Their wave-lengths are between 5774 and 5390. It is such a distinct absorption spectrum that it is possible to determine with certainty that the fluid actually contains a certain substance, even though the microscope might fail to give sure proof. Blood—human blood—that was what those stains were."
He paused. "The spectra of the blood pigments," he added, "of the extremely minute quantities of blood and the decomposition products of hemoglobin in the blood are here infallibly shown, varying very distinctly with the chemical changes which the pigments may undergo."
Whose blood was it? I asked myself. Was it of some one who had visited the tomb, who was surprised there or surprised some one else there? I was hardly ready for Kennedy's quick remark.
"There were two kinds of blood there. One was contained in the spots on the floor all about the mausoleum. There are marks on the arm of Dana Phelps which he probably might say were made by the teeth of my police-dog, Schaef. They are human toothmarks, however. He was bitten by some one in a struggle. It was his blood on the floor of the mausoleum. Whose were the teeth?"
Kennedy fingered the now set impressions, then resumed: "Before I answer that question, what else does the spectroscope show? I found some spots near the coffin, which has been broken open by a heavy ob-
