Academy and Tamiya 1/72 P-51 Mustang

KIT #: 1667/60749
PRICE: $12.50/17.50   MSRP
DECALS: Three Options in each kit
NOTES:  Apparently from the same molds.


            Who  me? Model a mustang?  Model multiple mustangs? Me? Who likes only trainers, utility transports, experimentals, and other also-rans? Chalk it up to age.  Merlin and Mach II were wearing me down and I needed a dose of instant gratification.  Besides -  what sort of a US A/C collection wouldn’t have a P51?

            I used to have several.  Revell’s 49 center was one of the original batch I snatched off the shelves of the Patch Barracks PX for only 26 cents apiece.  Those got me into adult modeling just about 40 years ago.  A prop and a few wheels are all that’s left of it, rattling around in the spare parts box.  If I recall correctly, it wasn’t a half bad rendering. There was a Matchbox P51-D sitting for many years in the Secret Model Airplane Museum, brush painted with the original Testors silver that never, ever dried. That and its Grand Canyon panel lines sent it to the scrapper when it fell behind the generally elevating  level of modeling. Then there was a Monogram B-variant, also from the half-buck era, setting perfectly acceptable in its unpainted OD plastic.  No idea what ever happened to it.

            That was actually the kit I was looking for.  Those ancient Monogram classics took all of an afternoon to build and paint into a fine model.  No sanding, no filling, no filing, no fiddling. They’ve been re-issued, by Accurate Miniatures I believe, at a whopping $30. I figured if I was going to shed big bucks I might as well see what  the high rollers roll out.  Traipsing through Squadron Shop’s e-catalogue I settled on the razor back from Academy and the bubble top from Tamiya. 

            MM already has reviews of these kits, and many photos on the Mustang Readers’ Page, but I thought I’d add my own tuppence ha’penny, particularly to cover the subject of color, which has been batted around on the Readers Forum.  Rather than, as usual, cranking the pictures through Photoshop to fix exposure levels, I’ve sent along raw shots taken under various lighting sources, most less than optimum, to demonstrate how drastically hues change in different viewing conditions. 


            Though the two companies are not known to be connected, both kits obviously came from the same molds, or at least the same mold makers.  Many parts are identical.  The parts breakdown and engineering are identical. Scribing, alternate choices, wheel well detail … all point to a common provenance, though some differences are present. Notably, the B model fuselage is laterally split aft of the cooing scoop, as if another variant was planned with a different empennage. (Editor's note: Academy has long been suspected of pirating other people's molds and changing the kit the required 15% to circumvent copyright laws)

            As far as quality goes, nothing is left to be desired in the molding, engineering, and fit. Both kits offer choices of canopies.  The B can be fitted with the bird cage or Malcolm hood. The D model also has two bubble tops, one supposedly slimmer for late or post-war models, though even under magnification the difference is  hard to discern. A selection of exhaust stacks is also offered with each kit, so careful photo research is needed to model a particular subject other than the ones in the instruction sheets, for which specific parts are referenced by number.

            As for accuracy, unless something seems drastically amiss, I don’t generally get out the micrometer and calipers.  One item did need checking.  The WEB waves are awash with weeping, wailing, and great lamentation over the accusation that P51-B models are kitted with the wrong wing; that these aircraft had a straight  tapered leading edge, unlike the “cranked” wing of the D variant, which had a pronounced change in taper near the root. I located a clear overhead view of a B Mustang (the URL for which has naturally disappeared from my scratch pad) which showed that the hairplane hexsperts  as usual are wrong.  The planform clearly shows a double taper  which matches that of the model, and frankly is indistinguishable from the D variant wing. If it’s not exact, it’s close enough for me.


            As always  starts with the cockpits, which is the only  point worth mentioning. Because seats and such can be inserted through the opening, these are best left to the post-painting stage.  The Academy B  featured complete, detailed, and well proportioned furnishings. The Tamiya D, however, was equipped with a seat and backrest assembly sized for a shrunken pilot.  Scratch building skill were not overly taxed by forming a new one out of bits and pieces.

            The Academy aft cockpit floor is slightly narrower than the fuselage width.  Resist the temptation to squeeze and glue, lest you end up sanding and re-scribing the now overly wide tail section. Other than those two points nothing is notable about the assembly, other than the lack of need for filler, even on the  often troublesome radiator scoop lip.


            NMF mustangs are boring, and impossible to mask without resort to $10-a-bottle Alcald, with all its foibles and prerequisites. O.D. over neutral gray is sexy .  Which brings up the point about color recognition.  The experts, once again, are always eager to point out that the shade in a particular photograph is not quite green enough, or too much purple, or a lemon yellow rather than a cadmium yellow.  Forum threads further are awash with angst over whether the federal standard gray in gloss differs from federal standard gray in matte.

The un-tweaked photos show what IMHO is the foolishness of all this fretting. The same paint under different light appears to have been poured from different cans.  Khakis and olive drabs are particularly prone.  In dim light they are nearly black. Under fluorescents they turn cyanotic. Close-up flash morphs them into desert pink, and the dark blue of the national insignia into robin’s egg pastel. 

Add to that uncertainty the inability of the photo medium, electronic or chemical, to capture and then display precisely what the eye perceives, not that all eyes equally  perceive. Additionally, there is the factor known as “scale effect”.  Even under identical light, large areas of color appear different from small swatches. The explanation for scale effect might best be left to experts in optical physics, or psychology; and while they’re at it, would they please explain why a “harvest moon” on the horizon looks ten times the size of  a mid-month moon high in the sky?

Having vented my spleen, also my pancreas and gall bladder, concerning obsessions with absolute color accuracy, I will nevertheless deprive no person of the pleasure of performing a fandango upon his Focke-Wolfe should he be tortured by thoughts of having doused it with  the wrong RLM (whatever that stands for).

I should also defend myself against inevitable accusations of gloss negligence, i.e. slathering in Future airframes that historically should be non-specular. In the words of Bonaparte:  “Pas de lieu rhone que nous”.  Or, you play with your toys the way you want; I’ll play with mine the way I want.

By way of overkill I might add:

If he can be shiny, I can be shiny.


            Not much noteworthy in this department. As mentioned, flight deck furnishings can be added through the cockpit opening; and the under-size D-model seat needed to be scratch built. A recent Off  Topic Forum thread concerned doubts regarding  the need to study algebra and other math subjects.  Permit this opportunity to demonstrate the applicability of algebra to life’s  most pressing problems: how big should be the part?  An exhaustive survey of all chairs in the domicile indicated a universally accepted tush width of 18 inches. Reducing that to 1/72   scale required resort to the formula for figuring proportions.  First establish the legend:  Let X = width of pilot’s seat in 1/72 scale. Therefore:  X is to 18 inches as 1 is to 72; or:   X / 18”  =  1 / 72

Cross multiplication yields:       72  X  = 18”  or  X  =  18 / 72

Resorting to Microsoft’s calculator accessory we find that  X = .25  or  ¼ inch

And, sure enough, a scratch-built seat one quarter inch wide and proportionately tall fit and looked absolutely perfect.

            There you have it, Folks; proof indisputable of algebra’s indispensability.  Stay in school, Kiddies.  And don’t let the dog eat your homework.

            I’ve already mentioned that both kits offer a choice of transparencies, which  can be posed open or closed.  Al alternate birdcage on the B has a separate side panel, but its thickness looks out of place.  A properly scaled window can easily be made from a snippet of acetate, but  I prefer all canopies closed.  Interiors in 1/72 scale are seldom convincing. Besides, it’s hard enough to keep clean the outside of a model, without having to worry about schmutz getting inside.


           A fellow can easily get spoiled sticking to these high quality kits.

Joel Hamm

March 2010

If you would like your product reviewed fairly and quickly, please contact me or see other details in the Note to Contributors.

Back to the Main Page

Back to the Review Index Page