484 
BULLETIN OF THE BUREAU OF FISHERIES 
bottom throughout the season in the Grand Manan Channel and locally in the Bay 
of Fundy. 
These several factors unite to make the coastal zone east of Penobscot Bay on the 
whole a more favorable environment for diatoms in summer than any other part of the 
gulf except Georges Bank, to be discussed later (p. 485). Since this theoretic 
generalization corresponds with the quantitative distribution of diatoms as actually 
observed during the warm months, the factors just mentioned are probably the chief 
ones which explain the persistence of rich flowerings of diatoms in abundance in the 
Mount Desert region and in Passamaquoddy Bay throughout the summer, con- 
trasted with their exhaustion in the Massachusetts Bay region by early May. I 
have not been able to trace the dependence of particular flowerings on physical 
or chemical conditions in the sea water more closely than this. 
Our failure to find diatoms in as great abundance between Mount Desert Island 
and Grand Manan as the flowerings farther west, on the one hand, or those reported 
by Fritz (1921) at St. Andrews at the mouth of the St. Croix River, on the other, is 
puzzling, for this section of the coastal zone not only receives a considerable influx of 
land water from several streams that may be expected to be rich in dissolved food- 
stuffs, but there is a dominant outflow along it from the Bay of Fundy. 
No part of the gulf becomes uninhabitable for diatoms even when the water 
becomes warmest and most stable and flotation most difficult. On the contrary, 
certain species then reach their maximum development, as an example of which the 
summer flowerings of Asterionella and Skeletonema will serve (pp. 431, 448). The 
latter, as it occurs in Massachusetts Bay, is especially interesting because the dates 
when an abundance of Skeletonema has been recorded in 1915 and 1922 (early 
autumn and late summer, respectively; p. 476) follow so closely the rise in the con- 
centration of silica recorded for late June in 1921 (fig. 134) as to suggest that it is the 
accumulation of silica taking place during the late spring and early summer (when 
there are few diatoms in that region) which makes the water there able to support 
the autumnal flowerings of Skeletonema. 
The general scheme of circulation in the gulf (with the water from the rivers 
tending to swing westward and to hug the coast line during most of the year, as 
shown by the distribution of salinity) is a sufficient explanation for the fact that the 
vernal flowerings of diatoms of its inner parts appear first close in to the land and 
attain a greater abundance and endure longer there than over the deep basin. The 
contrast in this respect between the coastal zone and the offshore banks, on the one 
hand, and the central deeps of the gulf on the other, simply reproduces on a small 
scale that between coastal or neritic waters and more oceanic regions in general. 
The gradual expansion of the diatom flowerings offshore from the land out over the 
central part of the gulf, where it does not reach its maximum until early May (p. 388), 
follows the offshore dispersion of the spring freshets of land water with their load of 
nitrogen, phosphorous, and silica. 
It is in just such areas as the open basin of the Gulf of Maine, where the tran- 
sition from a state of free vertical circulation in early spring is sudden to one of very 
pronounced vertical stability in summer, when the supply of nitrogen and of phos- 
phates from the deeps is thereby prevented, and where the silica content of the 
