Dear Michael,

I have a question regarding the output from the Anisotropy function from the BoneJ software. Forgive me if it seems basic but I want to double check the format and significance of the eigenvector and eigenvalue output.

For example, for one of my bone cubes I get the following output:

```
Fabric tensor vectors
||0.264| 0.288|-0.920||
||0.122|-0.957|-0.265||
||0.957| 0.042| 0.288||
Fabric tensor values
||0.010|0.000|0.000||
||0.000|0.024|0.000||
||0.000|0.000|0.047||
Image DA tDA V1,1 V1,2 V1,3 V2,1 V2,2 V2,3 V3,1 V3,2 V3,3 D1 D2 D3
cub102 0.786 4.676 0.264 0.288 -0.920 0.122 -0.957 -0.265 0.957 0.042 0.288 0.010 0.024 0.047
```

Regarding the eigenvectors, am I right in assuming that the vectors are arranged by column. Thus, in this example, the first fabric tensor vector direction is defined by the x,y,z coordinates V1.1 0.264, V2.1 0.122 V3.1 0.957 (as opposed to being defined by the values moving across each row e.g. V1.1 V1.2 V1.3).

Following on from this, am I right in assuming that each column i.e. eigenvector (or row if I am wrong!) corresponds to its associated eigenvalue i.e. eigenvector/column 1 is associated to the first eigenvalue, column 2 to the second eigenvalue and so on?

Lastly, regarding the eigenvalues, is the largest eigenvalue (0.047 in this instance) associated with the main principal direction i.e. the main trabecular direction? Or is the largest eigenvalue telling me that there are more boundaries (and therefore smaller intercept lengths) in this direction and thus it actually relates to the shortest principal direction?

Thanks for your time and for this brilliant programme!

JD