As Latin grew less important to general education, however, botanists have had to seek help from other quarters in writing these diagnoses, but because of the independent evolution of botanical Latin for its special purposes:
... when a botanical author thanks a professor of classics for providing a Latin description, this is usually in bad or at any rate unconventional botanical Latin... (W.T. Stearn, Botanical Latin, viii.)The use of Latin was codified in the International Code of Botanical Nomenclature - a set of rules for keeping the profusion of scientific names in order. It was preserved for both tradition's sake and also because taxonomists have usually argued that a common language is needed to keep communication possible amidst a profusion of vernacular languages. The parallel Zoological Code, however, did away with Latin descriptions a long time ago.
New editions of the Code are voted upon at meetings of the International Botanical Congress, and at the latest meeting in Melbourne, Australia, this tradition of botanical Latin may have met its demise (Nature news and editorial). Attendees have voted for several amendments to the code (full list of proposals here), including dropping the requirement for botanical Latin in new descriptions, and allowing the publication of new species names in electronic journals. The full congress will have to ratify these amendments before they become permanent.
The demise of botanical Latin might have been just a matter of time, but the issue of electronic publication is a relatively new one. Electronic-only journals have been proliferating, but for taxonomy, which is concerned with permanence and record-keeping rather than rapid communication of information, they have distinct disadvantages. There is no guarantee that a server crash or corrupted computer file may not wipe out a species description forever, resulting in a book-keeping nightmare when species names have to be revised or updated. The zoological Code, for example, still requires that a number of hard paper copies be sent to libraries around the world as an insurance against that possibility. This is what the Raffles Bulletin of Zoology, which is published in Singapore as an online journal, does to fulfill the requirements of the code.
So does this mean that William T Stearn's classic reference book, Botanical Latin, will become obsolete immediately? Not necessarily. A big fraction of the historical botanical literature, especially the earliest works from the time of Linnaeus onwards, are written exclusively in Latin and no translations exist for them. We also owe much of our technical nomenclature to Latin and Greek. Botanical Latin may be newly dead, but it still lives on.