Lemmatization fails on CAPITALS

Fresh download of MBSP 1.4 from Github today:

>>> MBSP.lemmatize("The cats were sleeping.", tokenize=True) 
u'the cat be sleep .'

For capital letters, only the first word works:

>>> MBSP.lemmatize("CATS WERE SLEEPING.", tokenize=True)
u'cat WERE SLEEPING .'

Other parts of MBSP have the same problem -- first word works, the rest fail:

>>> MBSP.parse('EATING PIZZA WITH A FORK.', lemmata=True)
u'EATING/VBG/I-VP/O/VP-1/A1/eat PIZZA/NN/I-NP/O/NP-OBJ-1/O/PIZZA WITH/IN/I-PP/B-PNP/O/P1/WITH A/DT/I-NP/I-PNP/O/P1/A FORK/NNP/I-NP/I-PNP/O/P1/FORK ././O/O/O/O/.'

>>> MBSP.tag(string, tokenize=True, lemmata=True)
u'CATS/NNS/cat ARE/VBP/ARE SLEEPING/NN/SLEEPING'

Sentences with initial capitals on non-initial words are handled correctly:

>>> MBSP.lemmatize("The Republicans were sleeping.", tokenize=True)
u'the Republican be sleep .'

Cheers,
David

Start by reproducing the MBSP.lemmatize, MBSP.parse, and MBSP.tag examples with uppercase input. Trace the lemmatization and tagging entry points to determine why only the first uppercase word is processed. Done means uppercase words throughout a sentence are handled consistently, while the shown mixed-case behavior remains correct.