This is simply (as inha said above) what minimizes the free energy of the system. Above a certain field, it is energetically cheaper to make little tubes of normal material within the SC that each allow a flux quantum through than to exclude the field entirely. The density of lines is determined by the applied field, so that the average flux density through the SC is equal to B outside (far away).
Marlon, you could look up the original paper by Abrikosov, if you can read Russian.
It is : A. A. Abrikosov, Zh. Eksperim. i. Teor. Fiz, 32, 1442 (1957)
I believe there's a translation at Soviet Phys. - JETP 5, 1174 (1957)
The concept of the fluxoid was first proposed by London. See F. London, Superfluids, vol. I, Wiley, New York (1950), p. 152
There's a nice review article by Tinkham, which I can not find now...will look for it.