Our main result that bilayer structure does not change with
mild dehydration down to 
suggests that the older neutron diffraction results
(Buldt et al., 1979) which were obtained for 
, should be valid for fully
hydrated 
phase DPPC. We here
show two ways that those results can be used to obtain 
. The
first way to use the neutron results employs Eq. 7,
but instead of using electron densities to obtain
, the differences in the locations of various molecular
groups between the fully hydrated gel and 
phases are used.
The headgroup labels at the 
, 
and 
positions yield 
2.6, 2.9 and 2.6 Å, respectively.
The hydrocarbon chain labelled at the 
position yields
Å. Only the glycerol label GC3 is out of line
with 
Å. Discounting this latter result and averaging
the other values gives 
Å and a value 
Å
using
Eq. 7. The second way uses the results for the positions of the
4th and 5th methylenes, 
Å
and 
Å
(although it may be noted that these
two spacings are mutually inconsistent because their difference
exceeds the C-C bond length of 1.54Å). We take the average
spacing of 11.4Å to include, on average, all the methylenes with carbon
number 5 or greater and all the terminal methyls. Assuming that the
volume of each terminal methyl is twice the volume of each methylene,
one has
where 
Å
(Nagle and Wiener, 1988).
