That agrees broadly with Louis Meulstee's "Wireless for the Warrior", these exhaustive tomes are generally right and well researched from the Royal Sigs archives in Blandford.

To précis a small part what is a huge reference work on the 19 set alone;
The DoND authorised production in the summer of 1941 and four Canadian engineers from Northern Electric were sent to England for design transfer. The design was later shared with the Canadian Marconi Co. and RCA Victor; they were tasked to produce 250 sets each through 1942 and the production peaked at 1000/mth in summer '42.

Early '42 the US Sigs Corps became interested and had an exact copy of the Canadian set produced to fill lend-lease orders; there is no evidence of US MkIII ever being made.

We must presume these were MkII sets since in the late summer of '42 British MkIII sets were supplied resulting in a Canadian prototype MkIII Oct '42; the previously mentioned suppliers made the change-over in May '43 but Northern Electric continued with the MkII for 2 months more to clear up orders.

There are minor detail circuit changes between the British and Canadian sets and large component differences since local sourcing was used, I suspect to be be sure of the origin you'd need to compare internally side by side and knowing the typical electronic component makers it would be obvious. eg. Canadian/Australian sets will not have Hunts capacitors, but beware some UK components were sourced in the US; moreover Pye here would be a cottage industry compared to the Canadian factories so the chances of finding a UK set are probably quite slim. Uniquely, the Canadains made a 24V supply for the 19 set too.

The Australian 19 set stemmed from the 1941 intention to raise an armoured division using locally produced vehicles, the armour initially supplied from the UK and US came with the 19 set MkII already fitted and the development of the 19 MkII (Aust) continued from there with the Australian AWA Pty. The first run was done in Feb '43 and a second in May '44. There are no references to any Australian MkIII production and even so the Aust version of the MkII is very noticeably different from all the others.

This version has a different number of valves and of different types, some switches are replaced with rotary types and the front panel is considerably changed, however, the set is mechanically and electrically interchangeable as a radio. The Aust. version was also far better environmentally protected, probably since the climate there demands all compnents, even for domestic use, to be rather different from the UK/Can/US types and so this set was very successful in the tropical environment whereas the other manufacture suffered alarming breakdown rates in the tropics. Owing to the initial design of the radio and subsequent widespread use, it was not possible during the war to redesign the whole thing into a sealed case.

There are no production figures available for Aust. production, however, known existing serials indicate over 600 were made.

There is a further complication, Australia re-worked a large number of 19 sets for Korean war use and these were re-designated MkII (Aust)/2. Generally these are found to be dated "R1952" and usually British MkII units.

R.