[ 100 ] 
little fpheres derive their origin ? my an Twer would 
be, without hefttation, from the polypi of the coral. 
And the reafon is, that, if thefe polypi produce their 
eggs, as will be fhewn in the fequel, covered with 
fuch corpufcles, wemayjuftly infer, that corpufcles 
of the fame nature, where-ever they are found, are 
formed by the fame polypi. 
§ 22. To this white pellicle is attached the bark 
of the coral (Fig. D. /, t , F, r, s y ) which is foft, of 
a vermillion-colour, or of a brighter colour than the 
coralline fubftance. It is formed of very fine mem- 
branulce y or net- work ; to which are annexed, and 
reciprocally fattened, the red globular corpufcles, 
which caufe its deeper colour. It is along this bark, 
that cylindrical veflels (Fig. D. t , t y t , Fig. E. /, Fig. 
F. n } ) are obferved to run lengthwife of the coral ; 
which appear by the microfcope to be parallel to each 
other, and out of which ifliie laterally other veflels 
infinitely fmall, ( E . t , t, t y ) which have a communi- 
cation with the above-mention’d membranula . The 
ufe of thefe veflels is to give nutriment to the coral, 
by means of a milky juice, which they contain. 
§ 23. The furface of this bark is flippery arid 
uneven, when the coral has been juft fifhed out of 
the fea ; fomewhat raifed in fome places, in others 
more deprefled and flatted. 
§ 24. Moreover, there are obferved in feveral 
parts of the faid bark little tubercles or prominences, 
(Fig. A. J,) which may be feen even without a mi- 
crofcope. Thefe tubercles are pretty large at their 
bottom or bafis, and round (Fig. /, n , n y ) grow fome- 
what narrower towards their upper part (0), and 
terminate in a lip of fome thicknefs, regularly divided 
into 
