Set here the direction autonomy, its relation with the atoms,
and which directions of the crystallographic family should be created.
Autonomy
When 
Autonomy is 
None, rotating, moving, scaling the
direction corresponds to rotating, moving, scaling the parent object. 
For crystallographic directions, the parent object is the cell containing 
the relevant crystallographic information. For atomic directions, the 
parent object is the first parent common to the two defining atoms. Atoms
in the 
thickness region are linked to atomic directions.
When 
Autonomy is 
Partial the crystallographic or atomic
direction can be rotated, moved, scaled, independently of its parent.
Atoms in the 
thickness region are linked to atomic or
crystallographic directions.
When 
Autonomy is 
All the crystallographic or atomic
direction can be rotated, moved, scaled, independently of its parent.
Atoms in the 
thickness region are copied to atomic or
crystallographic directions.
Thickness
The line representation of the direction is expanded as much as 
possible, limited by the cell volume, in crystallographic directions, 
and by the atoms within a 
Thickness range of the direction
defined by the two atoms, in atomic directions.
Node
A direction can be defined indicating explicitly the coordinates of a node 
where the direction passes. For each direction family, there is a direction
passing through the origin node and as nodes are equivalent, it follows
that for any node, there is a direction of any family passing through there.
By default, only the direction passing through the origin is represented,
with node coordinates 
o1, 
o2, 
o3, 
o4, equal
to 
0, 
0, 
0 and 
000, respectively.
o1, o2, o3
These entries provide the coordinates of the node in the lower-left
corner of the cell where the direction passes, calculated with conventional
or primitive cell vectors. When the lattice is primitive or vectors are
primitive, this corner node becomes the place where the direction passes.
The vectors used for the node coordinates, 
Conventional
or 
Primitive, are those used to define the plane indices,
in the 
Type page, of the first dialog.
o4
When the lattice is centered and vectors are conventional, a fourth
coordinate 
o4 is needed to point the centered node where the direction
passes. By default, 
o4 is 
000, so no change is introduced. 
When the cell lattice is primitive 
P or the vectors defining the 
node are primitive, that is the only possible value for 
o4. 
For I, C, F, R centered lattices, 
o4 can also take the values:
I: 111
C: 110
F: 110, 101, 011
R: 211, 122
corresponding to the numerators of the inner node coordinates,
(1/2 1/2 1/2) for 
I lattices, (1/2 1/2 0) for 
C
lattices, (0 1/2 1/2) (1/2 0 1/2) (1/2 1/2 0) for 
F
lattices and (1/3 2/3 2/3) (2/3 1/3 1/3) for 
R lattices.
Range
A set of directions to create can also be defined automatically, with two 
plane families belonging to the direction zone axis. Defining a range for 
each plane family, the intersection of the two ranges of planes defines 
an infinite volume, aligned along the direction orientation. All directions
of the family that fall inside this volume should be created. For directions 
[001], for example, the planes paralel, belonging to the direction axis, 
can be for example (100) and (010), as determined by the Weiss equation: 
hu + kv + lw = 0. For a cubic primitive lattice cP, a range (0 to 1) for 
planes (100) and a range (0 to 2) for planes (010) results in a volume 
containing 2x3 = 6 crystallographic directions [001]. To select a range 
of directions, press 
Range, to open a second level dialog.
Family
For each plane family, enter the indices 
h, 
k, 
l.
Each set of indices must obbey the Weiss zone equation, when combined 
with the direction indices, otherwise an error is shown.
Each plane of a family (h k l) intersects the lattice in n/h n/k n/l, where 
n = 0 means the plane passing through the origin and n = 1 is the usual
representation of the plane closest to the origin. The range of planes to consider
is defined by the values entered near to the buttons 
Start and 
End, 
describing the initial and final values of n. For example, setting 
Start 
to 
-1 and 
End to 
1 defines a range with 3 planes, 
intersecting the axes in: 1) -1/h -1/k -1/l; 2) 0; 3) 1/h 1/k 1/l.
Pressing the 
Start button, the entry is disabled and GAMGI considers
automatically all the planes from the beginning of the cell volume to the
final plane specified. Pressing the 
End button, the entry is disabled
and GAMGI considers automatically all the planes from the end of the cell volume
to the first plane specified. When both buttons are pressed, GAMGI considers
automatically all the planes from the beginning to the end of the cell volume,
so all directions inside the cell volume will be created.
After entering the plane ranges, pressing 
Ok saves the data,
closes the second level dialog, and disables the node information in
the first level dialog. Pressing 
Cancel, the current data in
both dialogs is maintained and the second level dialog is closed.
Pressing 
Node in the first level dialog, removes the second 
level data, closes the second level dialog, enables and initializes 
(if empty) the node data.
The vectors used for the node coordinates, 
Conventional
or 
Primitive, are those used to define the plane indices,
in the 
Type page, of the first dialog.
When adding information in the second level dialog, the direction indices
and the cell must have been entered before, so GAMGI can check if the 
information is correct. For the same reason, when 
Cell, 
u,
v, 
w, or 
Vectors changed in the 
Type page, 
all the information in the second level dialog is automatically discarded, 
as it might be wrong.