According to the 1905 Annual Report of the Navy Department, as of October 1, 1905 Naval shore stations at Portsmouth, New Hampshire, Boston, Massachusetts, New York, New York, Washington, D. C., Norfolk, Virginia, Key West, Florida, and Mare Island, California were all broadcasting time signals daily at noon.
 
The American Jeweler, October, 1912, page 411:

THE  FIRST  WIRELESS  TIME  SIGNAL.

    Captain J. L. Jayne, United States Navy, Superintendent United States Naval Observatory, Washington, D. C. writes:
    "I notice in your number for August, 1912, in an article entitled "The First Wireless Time Signal," that the claim is set forth that the first time signal was sent from the Marconi station, from Camperdown, Nova Scotia, in 1907. As a matter of fact, the United States Navy anticipated this by over two years. The fact that arrangements for such signals were being made was announced in the "Notices to Mariners," published by the Hydrographic office in November, 1904, and signals were actually sent out in January, 1905. The navy has been regularly sending out radio time signals from some of its coast stations since that date. It is possible that our work suggested the idea to our northern neighbors. I hope you will correct the impression that the Camperdown station got ahead of us. The facts that I have given above are matters of record in the Navy Department.
    "I read with pleasure the resolution of congratulation passed by the Jewelers' Association. I note, however, that the resolution spoke of the 'New Enterprise.' As a matter of fact it is only new in that the station is new and powerful and possibly some other details.
    "It is now expected that a radio (wireless) time signal will be sent out from the new station at Arlington at noon each day in exactly the same way as to grouping of sounds as those being sent out at present over the Western Union wires. It is probable, however, that a similar signal may also be sent out at night, on account of the fact that signals carry farther at that time. No decision has yet been made as to this point."
    The articles on wireless in the AMERICAN JEWELER during the last six months, and particularly the exhibit at Kansas City, have made the retailers take an attitude of the keenest interest. Much misinformation is being distributed and the following from the Washington Times is worth reprinting: It is not expected that the Arlington station will eliminate the Western Union time transmitting system; it will merely supplement it. There are many mechanical difficulties in connection with the subject which are yet to be solved. It is not known what wave length will be most effective at the wireless station. Furthermore, it is not certain just how far over the Alleghany Mountains the new station will be able to reach. It is expected that the flashes from Ft. Meyer will go 3,000 miles out to sea; but whether they will reach more than 1,500 miles to the west, or even that far, is yet to be shown. Therefore jewelers are warned not to purchase apparatus which may prove useless to them unless it can be tuned to the proper wave length.
    Synchronization of the railroad, postoffice and other public clocks throughout Germany by wireless telegraphy is also under way in Germany. This does not mean that the clocks are automatically synchronized, like a Self Winding Clock wired in the Western Union system in this country, but merely that wireless time signals will be sent out by the German imperial postoffice from the tower now being constructed at the town of Fulda. This tower will be over 300 feet high and the signals will be sent once each minute by a master clock closing the circuit of the radio transmitter.
    The London City Council has ordered that all public clocks must be synchronized, or in other words must be so regulated as to be within a reasonable distance of Greenwich time. For a number of years past it has been a standard subject of sport with the daily papers in London to make fun of the clocks. Recently one of the daily papers took a census of the public clocks in London with the result of showing a variation of 21 minutes, the clocks giving actual time being just over 3½ per cent of the total and on no street could more than two clocks be found to coinside. The difficulty has been owing chiefly to carelessness and the city council has ordered that this carelessness cease.