594 MR A. ANSTRUTHER LAWSON ON 



synaptic knot as a centre. The threads gradually thicken as synapsis proceeds, and the 

 length of the thread system is very much shortened. When fully contracted the 

 synaptic knot consists of much-thickened threads (constituting the spireme) which are 

 drawn so tightly together that their arrangement cannot be traced. The loops extending 

 from the contracted mass are at this stage thick and very conspicuous." 



Strasburger, Miyake, Gregoire, Berghs, and others have expressed similar inter- 

 pretations of this phenomenon, but the above extracts are sufficient to show how the 

 idea of synapsis has grown. 



Now, it is not the purpose of the present work to deny the existence of this stage 

 known as synapsis. I have observed it in many representative types of Algse, Fungi, 

 Bryophyta, Pteridophyta, Gymnosperms, and Angiosperms, and I am quite convinced 

 that it is a constant and normal phase in the nuclear-cycle. My interpretation of this 

 phase, however, is not in agreement with any of the above-quoted writers or with any 

 other interpretation which to my knowledge has yet been published. After a careful 

 study of many types, I have been convinced that during this phase known as synapsis 

 there is no contraction whatever of the chromatin substance ; that the condition so 

 often described as " contracted " is not in reality a contraction, but is subject to quite 

 a different interpretation. I have also been convinced that this so-called " contraction 

 stage " has nothing whatever to do with the blending or fusion of maternal and paternal 

 chromatin and consequently plays no immediate role in the process of chromosome 

 reduction. 



As these interpretations are so widely opposed to those of so many eminent 

 cytologists, I have delayed in publishing them until they had been confirmed by a study 

 of a wide range of forms extending from the Algse to the Angiosperms. This has been 

 done, and from the observations made I have no longer any hesitation in stating my views. 

 Now, while the following observations are confined to the microspore-mother-cells of 

 a single Angiosperm type, Smilacina, this was done for the convenience that this par- 

 ticular plant afforded in obtaining an unbroken series of stages of nuclear changes during 

 the heterotype mitosis. All of the main conclusions arrived at from a study of this plant 

 were later confirmed by an investigation of types from the Gymnosperms, Pteridophyta, 

 Bryophyta, and Algse. Smilacina is a particularly favourable plant for such a study, for 

 the flowers are quite small and very numerous in a close inflorescence. The entire 

 young inflorescences were taken and fixed in chrom-acetic acid without dissection. 

 After being imbedded in paraffin and sectioned, it was found that while at the base of 

 the raceme the tetrads were fully formed, at the apex the Archesporium was not yet 

 organised and that between these two extremes every conceivable stage of nuclear 

 activity was to be found. The advantages of having so many stages in a single section 

 are obvious — they allowed of a close and accurate comparative study of the different 

 stages in their natural sequence. 



The mature Archesporium showed the mother-cells fitting closely to one another with 

 the walls very thin and in the form of straight lines and sharp angles — leaving no room 



