408 
SUMMARY OF CURRENT RESEARCHES RELATING TO 
consists of two large tracheae, whose stigmata lie in the groove between 
the fourth coxa and the abdomen. Each foot receives two smaller 
branches from the main trachea, and these run down the femur and 
patella. The larger of the two communicates with the proximal spiracle 
of the tibia, the second passes down the tibia and reaches the distal 
spiracle. Herr Loman finds that in young Harvestmen (probably 
Phalangium opilio L. and AcantholopJms spinosus Bose.), the external 
openings on the tibia do not exist. The spiracles are post-embryonic, 
and arise from the tracheae within the tibia at a time when the appen- 
dages are rapidly growing and requiring more aeration. Perforation 
of the tracheal cavity probably occurs at eedysis. 
Spermatogenesis in Spiders.* — Herr J. Wagner gives a preliminary 
account of his researches. There are at first no cell-boundaries between 
the spermatogonia ; the nuclei do not show the usual karyokinesis, but 
are not amitotic. In the modification of the spermatogonium-nucleus 
into the spermatocyte-nucleus, the linin network gives rise to one linin- 
thread or to a few. The nuclei become excentric ; the linin with the 
chromatin granules passes into the half of the nucleus next the cell- 
periphery, and forms loops of similar length and direction. Archoplasm 
granules collect in the narrow space between the chromatin half of the 
nucleus and the surface of the cell, and in their interior lies a large central 
corpuscle. The nuclear membrane disappears as the linin and chromatin 
become aggregated. After the contraction of the archoplasm the greater 
part of the cytoplasm passes to that half of the cell. The number of 
central corpuscles in the spermatocytes of the first generation is 1-3. 
A true nucleolus of elliptical form lies always quite peripherally, and 
never within the linin-thread. In the first spermatocyte division the 
nucleolus divides either in the plane of the equatorial plate along with 
the chromosomes or outside these beside one of the spindle-poles. In 
the latter cases it passes out of the nucleus after the disappearance of 
the membrane. The achromatin-spindles do not disappear after the 
first and second divisions of the spermatocytes ; the second achromatin- 
spindle is formed independently of the first. 
Cell-division and nuclear division of spermatocytes are not quite 
contemporaneous ; multinuclear spermatides may first be formed. During 
the cell-division the round ZwisclienTcorperchen usually arises from the 
middle part of the spindle ; it may either migrate anew into the cell- 
substance or separate from both daughter-cells. The remains of the 
spindle always persist in the spermatids until they become spermatozoa, 
forming round bodies or seminal granules which are subsequently con- 
stricted off. During the modification into spermatozoa the nucleus 
becomes homogeneous and vesicular, the chromatin condenses into a 
plate which alone persists, and becomes elongated and spirally twisted. 
The axial thread of the tail is formed in the protoplasm of the sperma- 
tocyte or spermatid as a short rod, which unites with the above- 
mentioned chromatin plate, the function being marked by a little tooth 
on the plate. The mature spermatozoa of Tarantula are mobile. In 
the vas deferens both head and tail are so modified that a rod-like 
corpuscle without either head or tail is the final form. 
Zool. Anzeig., xix. (1896) pp. 188-90. 
