Acknowledged. Based on a sample size of two, it seems to me that google books doesn’t necessarily reveal earliest usages (which are generally spoken, AFAIK). But plausibly, they may show when a term has first “Arrived” as it were, and regardless they reveal a nice plotting of the trend in usage.
Google’s ngrams is a nice tool. (But see below: its reliability is unclear.)
Tough noogies first shows up in 1971. I see that Cecil used it in his 1984 publication.
http://www.google.com/search?q="tough%20noogies"&tbs=bks:1,cdr:1,cd_min:1900,cd_max:1984&lr=lang_en
Its popularity leaps and bounds:
http://ngrams.googlelabs.com/graph?content=tough+noogies&year_start=1900&year_end=2008&corpus=0&smoothing=3
tough tits:
Coming up with nada. Uh oh:
http://ngrams.googlelabs.com/graph?content=tough+tits&year_start=1900&year_end=2008&corpus=0&smoothing=3
But it does generate hits here. My earliest hit is 1962, not 1921 though. Puzzling.
http://www.google.com/search?num=50&hl=en&lr=lang_en&tbs=bks%3A1%2Ccdr%3A1%2Ccd_min%3A1900%2Ccd_max%3A1984%2Clr%3Alang_1en&q="tough+tits"&aq=f&aqi=&aql=&oq=&gs_rfai=
I guess ngrams is still in beta…