It had been settled that my edition of the Rig-veda should be printed at the Oxford University Press, and I found that I had often to go there to superintend the printing. Not that the printers required much supervision, as I must say that the printing at the University Press was, and is, excellent—far better than anything I had known in Germany. In providing copy for a work of six volumes, each of about 1000 pages, it was but natural that _lapsus calami_ should occur from time to time. What surprised me was that several of these were corrected in the proof-sheets sent to me. At last I asked whether there was any Sanskrit scholar at Oxford who revised my proof-sheets before they were returned. I was told there was not, but that the queries were made by the printer himself. That printer was an extraordinary man. His right arm was slightly paralysed, and he had therefore been put on difficult slow work, such as Sanskrit. There are more than 300 types which a printer must know in composing Sanskrit. Many of the letters in Sanskrit are incompatible, i. e. they cannot follow each other, or if they do, they have to be modified. Every _d_, for instance, if followed by a _t_, is changed to _t_; every _dh_ loses its aspiration, becomes likewise _t_, or changes the next _t_ into _dh_. Thus from _budh_ + _ta_, we have _Buddha_, i. e. awakened. In writing I had sometimes neglected these modifications, but in the proof-sheets these cases were always either queried or corrected. When I asked the printer, who did not of course know a word of Sanskrit, how he came to make these corrections, he said: “Well, sir, my arm gets into a regular swing from one compartment of types to another, and there are certain movements that never occur. So if I suddenly have to take up types which entail a new movement, I feel it, and I put a query.” An English printer might possibly be startled in the same way if in English he had to take up an _s_ immediately following an _h_. But it was certainly extraordinary that an unusual movement of the muscles of the paralysed arm should have led to the discovery of a mistake in writing Sanskrit. In spite of the extreme accuracy of my printer, however, I saw, that after all it would be better for myself, and for the Veda, if I were on the spot, and I decided to migrate from London to Oxford.