Michael Williams's paper on UTEC and Ferut mentions briefly the construction of a differential analyzer from Meccano parts similar to the one built by Douglas Hartree and Arthur Porter at Manchester at a cost of about $75. He says that this device was built by Trixie Worsley and that there were few indications of what it is used for. He concludes that it "was likely used only by Trixie Worsley to any extent". He gives the date of completion as September 1948, but a letter from my friend and colleague Jim Howland received after the paper was published would indicate that work continued on the differential analyzer for some time after this date. I give the following excerpt from this letter:
In 1951 I was asked to complete a Meccano model of a differential analyzer. Several parts were available, including two integrators, a plotter and an electric motor appropriate to driving the completed D.A. I had some difficulty in obtaining Meccano parts and spent a lot of the time scouring Toronto for the pieces I needed. I made two torque amplifiers out of brass in the machine shop at the Physics Department. These, operating on the principle of a capstan, provided a sufficiently strong output from the integrators to drive the other components of the D.A. The photograph in the Globe and Mail of December 15, 1951 shows the completed D.A. The test of the machine consisted in solving [an ordinary differential equation with initial conditions appropriate to plotting a circle.] This it did with an accuracy that surprised me, as the circle closed to within the width of the pen tip. I wanted to give Dr. Gotlieb the output together with a copy of the book The D.A. Draws a Circle (Earle Stanley Gardner, 1939). I could not find one however, so he just got the circle.
In my copy of the letter Jim added in a postscript that "I did not get my picture in the paper because, when the photographer came, I was downtown searching for more Meccano parts".
As a footnote on Mecanno Differential Analyzers we give the following quotation from an account of early calculating machines in New Zealand:
The most amusing (if not embarrassing) machine was made out of Meccano toy construction sets. This Meccano Differential Analyzer was originally built at Cambridge University in 1934 and used for various work in Britain. In 1950 it was brought out to New Zealand and was used by the Seagrove Radio Research Station. Later the Dominion Physical Laboratory employed it on problems associated with hydroelectric power station designs.