John Bryan said: "The exception being when the name ends in 'a' when an e is added, thus balansae from Balansa." Actually, what is being added in this instance is not an -e but rather the letter combination ae. And it's being added to the stem Balans- . The ae is the feminine singular genitive form. More about this below; it gets more and more curious. [In John's defense, it should be noted that the Rules express it the same way - they say add an -e). That advice obscures what is really happening.] John has quoted the rule as it applies to specific epithets; for generic names, one adds -ea to words which end in -a to form a generic name. The name John used, balansae, is (or can be) a specific epithet, but not properly a generic name. There are two aspects of this rule (actually it is Recommendation 60B) which are bizarre in my opinion. Let's get the easier one out of the way first. In the example used above, based on the name of the French botanist Benjamin (Benedict) Balansa, (do I need to say a male person?) the botanical names (both the generic name and the specific name) are feminine. In other words, the rules treat names ending in -a as feminine words. Never mind that in their language of origin there might not be any such gender implications. Since the name in question is a French name, and French does make such gender distinctions, enough said about this one. Now on to the second aspect of the bizarre nature of this recommendation. This is where doing things by rote leads us down the wrong path. The recommendation fails to distinguish between vowels and vowel sounds. The linguistic basis for the changes under consideration are based on the sounds involved, not the spelling. The spelling is irrelevant; but since the purpose of the rules is to codify spelling, things have gone awry. For instance, here's what the rules say Recommendation 60B 60B.1. When a new generic name, or subgeneric or sectional epithet, is taken from the name of a person, it should be formed as follows: (a) When the name of the person ends with a vowel, the letter -a is added (thus Ottoa after Otto; Sloanea after Sloane), except when the name ends with -a, when -ea is added (e.g. Collaea after Colla), or with -ea (as Correa), when no letter is added. Do you see the problem here? Say the name Sloane. As spelled, it does end in a vowel. But the vowel is silent. As pronounced, it ends in a consonant sound. In fact, it ends in the "n" sound. It should be treated as a word ending in a consonant. And how many syllables does this Sloanea have? Two? (say Sloane and add an a) Three? (say Slone + e + a ) Four? (slo-a-ne-a). (At this point it occurs to me that maybe I don't know how to pronounce Sloane. I assume it is pronounced as a one syllable word.) They are paying too much attention to the written form and are ignoring the underlying sounds. Nor is it hard to understand why: isn't the purpose of these names to stroke the vanity of the person honored? Wouldn't they complain if you re-spelled their name (misspelled their name)? Linnaeus wrestled with this problem, too. He coined the name Stranvaesia, based on the English name Strangeway, to honor a wealthy benefactor. My impression is that Linnaeus was making a good faith effort to latinize the name. No doubt the honoree complained, perhaps saying that his name was now unrecognizable or looked foreign. Whatever the case, Linnaeus soon dropped the practice of re-spelling names so that if pronounced as Latin they would approximate their sound in their language of origin. And for better or worse, that's the predominant way it's been done since. Spelling rules! So does doing things by rote! Jim McKenney