The little man bustled to a cupboard and brought forth a singular appliance, of his own manufacture, somewhat like a miniature vacuum cleaner. It had been made from a bicycle foot-pump, by reversing the piston-valve, and was fitted with a glass nozzle and a small detachable glass receiver for collecting the dust, at the end of a flexible metal tube.

"We will sample the dust from the outside first," said Thorndyke, laying the hat upon the work-bench. "Are you ready, Polton?"

John Thorndyke's Cases — Part 72

The assistant slipped his foot into the stirrup of the pump and worked the handle vigorously, while Thorndyke drew the glass nozzle slowly along the hat-brim under the curled edge. And as the nozzle passed along, the white coating vanished as if by magic, leaving the felt absolutely clean and black, and simultaneously the glass receiver became clouded over with a white deposit.

John Thorndyke's Cases — Part 72

"We will leave the other side for the police," said Thorndyke, and as Polton ceased pumping he detached the receiver, and laid it on a sheet of paper, on which he wrote in pencil, "Outside," and covered it with a small bell-glass. A fresh receiver having been fitted on, the nozzle was now drawn over the silk lining of the hat, and then through the space behind the leather head-lining on one side; and now the dust that collected in the receiver was much of the usual grey colour and fluffy texture, and included two more hairs.

John Thorndyke's Cases — Part 72

"And now," said Thorndyke, when the second receiver had been detached and set aside, "we want a mould of the inside of the hat, and we must make it by the quickest method; there is no time to make a paper mould. It is a most astonishing head," he added, reaching down from a nail a pair of large callipers, which he applied to the inside of the hat; "six inches and nine-tenths long by six and six-tenths broad, which gives us"--he made a rapid calculation on a scrap of paper--"the extraordinarily high cephalic index of 95·6."

Polton now took possession of the hat, and, having stuck a band of wet tissue-paper round the inside, mixed a small bowl of plaster-of-Paris, and very dexterously ran a stream of the thick liquid on to the tissue-paper, where it quickly solidified. A second and third application resulted in a broad ring of solid plaster an inch thick, forming a perfect mould of the inside of the hat, and in a few minutes the slight contraction of the plaster in setting rendered the mould sufficiently loose to allow of its being slipped out on to a board to dry.