Click on Analyse Results then on Journals and finally on the small square beside Phys.Rev.Letters.
The resulting plots show the number of citations and papers in Phys.Rev.Letters as a function of year..
Note that these numbers increased strongly until the early 90's, then fell off significantly. The numbers increased again even more strongly in the latter part of the 90's and we are now doing much better than ever before ! (There is of course a tail off for very recent papers, which have not yet been published long enough to be cited). This dip in ILL scientific impact corresponds with the 3-year reactor shutdown at the start of the 90's.
PRL citations, as well as those in Nature, respond rapidly to drops in ILL productivity because the papers tend to be very topical but are then less cited over time. Papers in other journals, especially review or technical journals like Nuclear Instruments and Methods, tend to be cited more regularly and do not reflect such productivity drops.