1902] CURRENT LITERATURE 75 



shaft more or less elongated, dependent probably on the depth of the primary 

 tuber. The top of the cylindrical body becomes expanded in various ways 

 and bears the first sex-organs. Finally the whole structure becomes very 

 irregular, and the whole upper part of the mature prothallium is green, 

 excepting the conspicuous archegonial necks. The antheridia and archegonia 

 seem to differ in no important particular from those of Lycopodium. The 

 embryo, immediately on its escape from the prothallium, forms the protocorm, 

 whose "pedicel " elongates and thrusts it to a safe depth (about 3""*). Crib's 

 claim that he sowed the spores of Phylloglossum and obtained a colorless 

 prothallium like that of Ophioglossum can only be explained on the hypoth- 

 esis that he saw merely the primary tuberous portion of the prothallium, for 

 the later and conspicuous part becomes even vividly green. On the whole, the 

 prothallium resembles that of Lycopodiiun cernuiim nearer than any other. 

 Bower's observations of the occasional branching of the strobilus, and of a 

 J leaf some distance below the strobilus, are confirmed; but this leaf was 



I* always a sterile sporophyll, and no transition between prolophyll and sporo- 



phyll was observed. Bertrand's claim that Phylloglossum is a reduction 

 form on account of its semiaquatic habit is controverted by the fact that the 

 plants grow better on a hilltop than on the slope, and Thomas never found 

 them in an actual swamp. The writer inclines to the view of those who 

 regard Phylloglossum the most primitive of living Lycopodiales. — J. M. C. 



Karsten, in describing the embryology of the Juglandaceae," emphasizes 

 features in which this group resembles the gymnosperms, and expreses his 

 belief that the angiosperms are derived from the gymnosperms, Gnetum being 

 the point of contact. The forms studied were Juglans regia, J, cordiformis, 

 /. nigra, Pterocarya fraxinifolia, Carya amara, and C tomentosa. In Jtig- 

 la?is cordiformis at the time of fertilization, and even later, the union of the 

 carpels is delayed, thus leaving the ovules exposed, so that the condition 

 hardly differs from that in those gymnosperms with orthotropous ovules, as 

 Gnetum. There is an extensive sporogenous tissue in the nucellus, but any 

 further development is usually restricted to a single cell in the median plane. 

 The megaspore mother cell may develop directly into the embryo sac, or may 

 first give rise to a row of three or four megaspores, the two upper of which 

 never develop, but the two lower seem to have an equal chance. Many 

 cases were found in which there were two embryo sacs in the median plane, 

 lying one above the other, sometimes touching each other at the ends, but 

 often separated by several layers of sterile cells. A case is figured in which 

 the sac nearest the chalaza had been fertilized. In Juglans nigra the ^g^ 

 cannot be distinguished from the synergids before fertilization. There is 

 probably no fusion of the polar nuclei, and if it takes place at all it occurs 



Karsten, Geo., Ueber die Entwicklung der weiblichen Bluthen bei einigen 

 Juglandaceen. Flora 90:316-333.//./^. 1902. 



