A felicitous but unproved conjecture may be of much more consequence for mathematics than the proof of many a respectable theorem.