The essence of the article is that John Crane was also hiding Rife's work. Instead of being consistent with dividing down the true Rife frequencies, he used random numbers. Examples:
Rife 478,000 is divided by 627 which is commonly used as 727 or 728, but 727.5 would be better.
417,000 hertz original Rife is divided by 521 to come up with 800.4 hertz, which is rounded to 800 hertz.
120,000 hertz (tetanus) is divided by 1950 to come up with 120 hertz exactly.
789,000 hertz is divided by 1195 to come up with 660.3 which is rounded to 660 hertz. etc, etc.
Now you also see why the Crane frequencies should be used to the decimal point if you want to be more correct. Also you don't hit the original Rife frequency when using a 50% duty cycle when Crane used an even number to divide an original Rife frequency. A 50% duty cycle only hits the odd number multipliers. Now you see why they often use other duty cycles in applying the Crane frequencies, (such duty cycles may be 71% , 78%, 88%, 90%, 93%, depending on what random digit Crane used to divide the original Rife frequency by).