322 Scientific Proceedings, Royal Dublin Society. 



of the other cells of the tetrad were still visible in Larix leptolcpis, sufficient to 

 demonstrate that the formation was linear ; but whether there were three cells 

 or four could not be determined. As would be expected, only one rnegaspore 

 is formed. On March 23rd the first division was seen. All through April 

 only the typical parietal layer was forming, such appearing even in ovules 

 gathered on April 28th. Once May set in the activity was rapid, so that the 

 only stage obtained on May 5th showed the endosperm so well formed 

 that it was impossible to say how it had grown in. On May 12th, though the 

 endosperm was still far from compact, archegonial development had well 

 begun and was quite normal, a superficial cell dividing to give an inner 

 archegonial cell and an outer neck cell. The rest of May was occupied by 

 the maturing of the whole gametophy te, which, and especially the archegonium, 

 merits perhaps a few words of description. 



Plate XVIII, fig. 8, shows the archegonium in longitudinal section. It 

 is relatively long and thin and recalls that of Pseudotsuga,. Lawson (15), 

 however, referring to a photomicrograph through a complete female game- 

 tophyte of Larix europea which appears in Thomson's well-known paper 

 on " The Megaspore Membrane " (27), says that of Larix seems thicker than 

 that of Pseudotsuga. Difficult as it is to see the difference when a comparison 

 is made with Thomson's photograph, a comparison with the fig. on PI. XVIII, 

 only emphasizes the general similarity of the archegonia of the two genera. 

 The archegonia of Larix europea are, however, certainly relatively wider. The 

 archegonia are always five in number (PI. XVIII, fig. 9). Though the jacket 

 cells are very marked all round the archegonium in fig. 8, the jackets of 

 neighbouring archegonia are almost in contact, as fig. 9 shows. The double 

 layer may be reduced to one, and this even to a thin membrane at times. 

 Lawson reports the same of Pseudotsuga, with which Larix also agrees in 

 its shallow archegonial chamber. There are one or two tiers of neck cells, 

 though one layer appears more common. The cells of each tier vary from 

 four to eight in number (PL XVIII, fig. 10). The cytoplasm at this stage is 

 markedly vacuolar and the nucleus apical. As already reported by Thomson 

 for L. europea and L. amerieana, the megaspore membrane, very definite and 

 thick at the bottom, thins out so markedly over the archegonial region that 

 at maturity no trace of it can be seen there. Pseudotsuga shows a quite similar 

 condition, and in both the condition is very much more definite than in the 

 other Abietineae so far described. (This is in interesting contrast to Pseudo- 

 larix (17), in which the megaspore membrane is quite distinct even at the 

 micropylar end of the endosperm.) So that taken all in all a description of 

 the female gametophy te of Larix leptolcpis might well be a description of that 

 of Pseudotsuga Douglasii. 



