282 
Petrocelis Hennedyi and Cruoria pellita. Here also it must be presumed that the 
tetrasporangia are formed without reduction of chromosomes. 
There are thus a considerable number of Cryptonemiales which differ with 
regard to the course of development from the typical diplobiontic forms. 
SvEDELIUS'), referring to the simultaneous occurrence of monospores and te- 
traspores in one and the same individual of Chantransia efflorescens, considers it 
not altogether impossible that future investigation of the cruciate tetrasporangia 
may show them to have been produced without reduction of chromosomes. Up 
to the present, however, no Floridea with such sporangia has been subjected to 
closer cytological investigation. The Swedish writer points out in this connection, 
that such sporangia are first divided by a transverse wall, and thereafter by two 
perpendicular partitions, which he considers would hardly fit in with a reduction 
division. It should nevertheless be borne in mind that we find, both in Arche- 
goniates and in flowering plants, cruciate sporangia as well as zonate sporangia, 
— though the latter, it is true, are more rare — and it seems not to be apparent 
that the formation of a cell-wall on the first division would preclude the reduction 
of chromosomes. As regards the zonate division, it has in several of the Coralli- 
paceæ been demonstrated with certainty that the three cell-divisions take place 
almost simultaneously, and that the nuclear divisions are completed before the cell- 
division sets in (see p. 273). It is hardly likely that there should be any difference 
in principle between the cruciate and the zonate division; among other reasons, 
because we find both occurring in the species of the genus Hildenbrandia, — which 
are doubtless very closely related — where the sporangia must also be presumed to 
divide without reduction of chromosomes (cf. also Lithothamnion Sonderi, fig. 137). 
If SvEDELIUS’ supposition were correct, it would involve either that the reduction 
division must take place by the division of the zygote nucleus, in spite of the 
presence of tetrasporangia, or that it never occurred among Cryptonemiales, since 
the tetrasporangia, as far as we know, here never divide tetrahedrically, but always 
by parallel or cruciate walls, often markedly inclined. The latter alternative would 
further imply that the cystocarpia were throughout developed by parthenogenesis, 
which is not in accordance with the actual facts, as, though fertilization has not, 
it is true, been cytologically demonstrated in any of these alge, which are furnished 
with tetraspores?), yet spermatia have at any rate been found attached to the 
trichogynes in Dumontia incrassata (see above p. 158), Polyides rotundus (THURET, 
Et. phyc. Pl. 38 figs. 14—18) and in certain Corallinaceæ (Choreonema Thuretii, SoLMS, 
Corall. Taf. III, fig. 4, Corallina mediterranea, Sons, 1. c. Taf. III, fig. 19). 
On the other hand, it must be presumed that reduction division may also be 
lacking in tetrahedrically divided sporangia, as cases are also known where such 
1) N. SvEDELIUS, I. c. p. 50. 
*) The fertilization has been cytologically demonstrated in Gloiosiphonia capillaris by OLTMANNS; 
but this Alga has usually no tetrasporangia. 
