The fire marshal leaped to his feet and bounded over to where Kennedy was seated. With one hand on Craig’s shoulder and the other grasping Craig’s hand, he started to speak, but his voice choked.
“Thanks,” he blurted out huskily at last. “My reputation in the department is at stake, my promotion, my position itself, my—my family—er—er—”
“Not a word, sir,” said Kennedy, his features working sympathetically. “To-night at eight I will go on watch with you. By the way, leave me those A. Spark notes.”
McCormick had so far regained his composure as to say a hearty farewell. He left the room as if ten years had been lifted off his shoulders. A moment later he stuck his head in the door again. “I’ll have one of the Department machines call for you, gentlemen,” he said.
After the marshal had gone, we sat for several minutes in silence. Kennedy was reading and rereading the notes, scowling to himself as if they presented a particularly perplexing problem. I said nothing, though my mind was teeming with speculations. At length he placed the notes very decisively on the table and snapped out the remark,
“Yes, it must be so.”
“What?” I queried, still drumming away at my typewriter, copying the list of incendiary fires against the moment when the case should be complete and the story “released for publication,” as it were.
“This note,” he explained, picking up the first one and speaking slowly, “was written by a woman.”
I swung around in my chair quickly. “Get out!” I exclaimed sceptically. “No woman ever used such phrases.”
“I didn’t say composed by a woman—I said written by a woman,” he replied.
“Oh,” I said, rather chagrined.
“It is possible to determine sex from handwriting in perhaps eighty cases out of a hundred,” Kennedy went on, enjoying my discomfiture. “Once I examined several hundred specimens of writing to decide that point to my satisfaction. Just to test my conclusions I submitted the specimens to two professional graphologists. I found that our results were slightly different, but I averaged the thing up to four cases out of five correct. The so-called sex signs are found to be largely influenced by the amount of writing done, by age, and to a certain extent by practice and professional requirements, as in the conventional writing of teachers and the rapid hand of bookkeepers. Now in this case the person who wrote the first note was only an indifferent writer. Therefore the sex signs are pretty likely to be accurate. Yes, I’m ready to go on the stand and swear that this note was written by a woman and the second by a man.”
“Then there’s a woman in the case, and she wrote the first note for the firebug—is that what you mean?” I asked.