Ann: Musket Drilling Update , page-27

Here is a very crude and simplified interpretation ("guesstimate") of the Musket deposit as of 27/05/2013. If you do read on, please note that this is not in any way a legit or ASX-compliant estimate of the Musket resource. This guesstimate was done in a few hours where normally should take multiple weeks with proper raw data and a "Competent Person" doing the estimation. This guesstimate is not categorized in any resource classification but only taken as an indication of the potential size of the current Musket deposit. I also feel I need to disclose I am a holder of ROX and have been for a very long time (years) so naturally I have a conflict of interest for this purpose. But if you do feel like seeing what a guy with a few spare hours did tonight, please continue on.

Below is me sharing my PERSONAL interpretation:

Here is how I did my guesstimate:
The grade shell is > 1% Ni (although most of the reported intersections are much higher in the 2-3% range but since the lowest reported mineral interval is 1.04% Ni, I am obligated to say 1%). The informing samples are simply the average grade of the mineralised intervals. There is no consideration of grade capping as my sample grades are only average grades reported in the announcements. Only one domain is used, and all mineralised zone is considered fresh rock with SG of 3.1 gm/cc. Variography was done and modelled (Nugget of 0.06, Spherical structure: sill of 0.175 and range of 110m) on the available samples for use in Ordinary Kriging. Anisotropic ratios are set at 3 and 3. Search ellipse is in the main direction of the variogram model and the main search distance corresponding to the modelled range. A minimum of 5 samples are required per estimate. Guesstimation was done with two passes. The first pass uses the above parameters and the second pass only filled blocks that weren't filled in the first pass. The search distance for second pass was expanded to 3 times the modeled range with the minimum number of informing sampled drop to 3 instead of 5 (due to low sampling density to the northern part of my model). Min block sizes are 5 x 5 x 2.5. For those of you that didn't understand the above paragraph, look up Geostatistics.

Here is the part I feel I gave Musket a bit more room:

The northernmost nickel intersection (MFEC037: 3m @ 3.13% Ni) stops at 7033994mN and I pushed it another 50m to 7034100mN (half way to MFEC038) as I felt there is good likelihood of the mineralisation continuing in that direction. Of course only drilling will tell but the current results suggest a good possibility. An addition of a few more drills holes could support this.

Also note that there are possible PGE credits to add to Musket so there could be more value there but will probably be offset by the effect of grade capping in the actual estimate done later on this year. But nonetheless, PGE is a significant 15% on top of Nickel for the high grade zone (see my post dated 27/03/14 12:12).

So at a 1.5% Ni cut-off, I interpreted Musket to be ~ 2.5Mt @ 2.1% Ni (54kt Contained Ni). At a 2% cut-off, I have 1.6Mt @ 2.3% Ni (36kt Contained Ni).

Below are my the results of my guesstimate:

Blue blocks= no nickel grade
Green blocks= > 1% Ni
Orange Blocks= > 2% Ni
Red Blocks= > 2.5% Ni

Please understand this is a very simple guesstimate without the proper assay results. The reported intervals are already smoothed (averaged) and the interpolation method also inherently smooths the estimate result so hence the lack of defined high-grade structure in the krig image.

PLease Please DYOR, this is my own interp. But I thought I would share as there are a few worked up on possible resources figures.

I'm looking forward to the additional drill holes at Musket, seems to be coming together very well. Patience is definitely required for this industry. Good luck to all!