612 DR A. ANSTRUTHER LAWSON ON 
spireme. This becomes even more convincing as the threads become more condensed. 
That all of the free ends observable in figs. 18, 14, and 15 are due to the cutting of a 
continuous coiled spireme by the microtome knife, cannot be maintained, for, as shown 
in fig. 15, the ends occur with great frequency in positions that are not at the surface 
of the sections. It should also be mentioned that throughout this entire series no 
feature that would even suggest a transverse segmentation was found. If further 
evidence were needed to show that the looping method of Farmer (1905) and Dicpy 
(1910) is not a feature of the prophase of Smilacina, it may be found in the series of 
later stages represented in figs. 17, 18, and 19. Here the free ends of the uniting 
chromosomes are clearly visible. The still later stages shown in figs. 20, 21, 22, 23, 
and 24 demonstrate how the pairing chromosomes coil and twist very tightly about one 
another as they become more condensed and shorter. 
In Kniphofia 1 was fortunate in securing a beautiful series of prophase stages, which 
confirm in all essentials the description given above of the manner in which the bivalent 
chromosomes are developed in Smilacina. In fig. 45, for instance, we have represented 
a stage soon after the close of the growth period. A number of free independent 
spiremes may be seen quite easily, and each one shows evidence of a longitudinal fissure. 
Fig. 46 represents a slightly older condition, where the spiremes are thicker and the 
longitudinal fission is less evident. In figs. 47, 48, and 49 it may be observed how the 
pairing and conjugation of the spiremes (somatic chromosomes) actually occur. It will 
be noticed that the ends of the two chromosomes, about to unite, come together first. 
This feature should not be mistaken for a looping or for a telosynaptic union. As shown 
in figs. 48 and 49, the union is clearly a lateral one, resulting in a coiling and twisting 
of the two chromosomes about one another, in exactly the fashion described above for 
Smilacina. 
In Aloe the very same conditions seem to prevail. Fig. 50 is drawn from a section 
taken at a time soon after the close of the meiotic growth period. It corresponds 
to the stage represented in fig. 15 of the Smelacina series and fig. 45 of the Kniphofia 
series. As in these other two plants, the conditions here show a number of independent 
spiremes each with a longitudinal fissure. The free ends of the chromatin threads may 
be seen without the slightest difficulty. The manner of their approximation and lateral 
conjugation in pairs is illustrated in figs. 51 and 52. Fig. 51 is interesting, because if 
the preceding stages had not been examined it might be mistaken for a looping. A 
close examination, however, will convince the observer that this effect is produced by 
the approximation of the ends of two chromosomes before the sides have come in con- 
tact—and not to the bending of a single filament. The characteristic coiling and 
twisting of the pairing chromosomes about one another is indicated in fig. 52. The 
resulting bivalent chromosomes continue to condense and shorten until the time of 
spindle formation. 
These observations on the meiotic prophase of Smilacina, Kniphofia, and Aloe, 
convince me beyond question that the chromatin never consists of one continuous 
