next message in archive
no next message in thread
previous message in archive
previous message in thread
Index of Subjects
> row. This was a light sandy-loam (as I learned much later) but Probably few care, but in the previous long post on this there's a silly arithmetical mistake: > 'So if you had a huge wasp SGO with 630 cells, the S/N improvement > would be twice that of a roach with only ~25 cells [sqrt(630) = 25]'. No - it wouldn't be twice that, it would be five times that. The '630 cells' is in the correct direction but is way overblown to make the case. To double the Signal-to-Noise ratio for the known 25 cells of the roach SGO where the signals converge in the CNS, you need to multiply by (2 squared = 4), to produce 100 cells. The square root of 100 is 10, the sqrt of 25 is 5, so the S/N improvement then would be ideally 10/5, a factor of 2, or 100% over the original, as the wording intended. Making 630 cells converge ideally would give you a 25/5 improvement over the roach, or a 5-fold increase in S/N, 500%, unnecessarily large for the argument being made. As indicated previously, it is not known how many sense cells are present in the pelecinid's SGO in its huge hind tibia -- so far no-one has looked. Steve
next message in archive
no next message in thread
previous message in archive
previous message in thread
Index of Subjects