Animal protein II

Red meat

SLOP1640 — week 4

Idris Fenn

The same structure, on purpose

Last week’s structure repeats here: anatomy first, then handling, then portioning.

  • beef is the primary material — the animal most students will have handled in the widest range of cuts
  • goat and lamb are the same argument applied to a smaller frame with a different fat distribution, not a separate topic

Worked muscle versus resting muscle

A cut’s texture is a direct record of what the muscle did in the living animal.

  • leg, shoulder, neck — worked constantly against load, dense fibre, high collagen (chewy, unrendered)
  • loin, along the animal’s back — did comparatively little work, far less connective tissue, finer grain
Beef cuts by muscle group, mapped to the anatomy this slide describes
A raw beef shank cross-section, a worked cut with coarse muscle grain and a visible band of connective tissue
Raw beef tenderloin cut into portions, a resting cut with fine, even muscle grain and almost no connective tissue

What follows from that

  • tender with minimal treatment
  • dries out rather than softens if treated the way a shoulder is treated
  • fat accumulates where the animal stored energy, not where it worked hardest — a related but separate logic from connective tissue

Goat and lamb, briefly

  • proportionally less external fat than beef of a comparable cut
  • smaller muscle groups — connective tissue is a larger share of any given cut’s mass

Treatment follows anatomy, not convention

  • high-collagen cut — needs time and moisture for collagen to convert to gelatin
  • low-collagen cut — gains nothing from the time a tougher cut requires
  • an anatomical answer, even though tradition usually arrives at the same answer by trial rather than by dissection

Unchanged from week 3

  • refrigerate promptly — within 2 hours, 1 hour above 32 °C (90 °F)
  • keep at or below 4 °C
  • treat any surface or tool that has touched raw meat as contaminated until washed
A hand holds an appliance thermometer inside a refrigerator, checking its reading against the food stored on the shelves rather than trusting the dial on the door

Portioning

Along, rather than across, the direction of connective tissue and fibre — this is what makes a portion cook predictably.

The same logic scales with portion size. A primal cut carries the least surface per gram; breaking it down for sale or for the kitchen adds cut surface, and cut surface is where surface bacteria take hold first. Carried to its extreme — grinding — that is why ground red meat keeps for 2 days against the 3–5 days a whole cut keeps (see Whole cuts versus ground, ahead).

Whole cuts versus ground

MaterialRefrigerated window
Beef, veal, lamb, pork (whole cuts)3–5 days
Poultry, fish (week 3)2 days
Ground/minced red meat2 days

Grinding mixes surface bacteria through the whole volume, so ground red meat follows the shorter window rather than the longer whole-cut one.

A whole raw beef standing rib roast, tied with butcher's twine, with only one cut face exposed and the rest of its surface intact
Ground beef heaped on a plate, its surface broken into many small fragments rather than one intact face

The danger zone, unchanged

The 4 °C to 60 °C danger zone applies identically to red meat as to poultry and fish.

Bright red, then brown: oxidation, not spoilage

A freshly cut surface is bright cherry red; refrigerated, it turns brown. Same pigment, myoglobin, in different chemical states — not different levels of safety.

  • oxymyoglobin — myoglobin freshly bound to oxygen, bright red, the colour of a cut just made or a package just opened
  • metmyoglobin — the same pigment further oxidised by air and store lighting, brown
  • USDA FSIS: this colour shift alone does not mean the meat has spoiled — an intact cut’s interior is often greyish from lack of oxygen, which is normal
A raw beef steak on a cutting board, its cut surface a bright cherry-red where it has been freshly exposed to air

What actually indicates spoilage

Colour records what a surface has been doing chemically since it was cut. It does not, by itself, say whether the meat is still sound. FSIS names three checks that do:

  • slime or tackiness on the surface — a texture change, not a colour change
  • an off odour — sour or otherwise unlike the neutral smell of fresh meat
  • the date on the label, read against how long the meat has actually been held

None of the three is colour. A brown cut with none of them is oxidised; a bright red cut with any one of them is not fresh, whatever its colour says.

Two ground-beef blends, per 100 g

Raw, from USDA FoodData Central — the same two blends, and the same figures, that this week’s tutorial calculator uses.

CutCalories (kcal)Fat (g)Protein (g)Iron (mg)FDC ID
Beef, ground, 85% lean / 15% fat, raw21515.018.592.09171796
Beef, ground, 93% lean / 7% fat, raw1527.020.852.33173110

Lamb leg and goat, per 100 g

Same source and convention as the previous slide. Both are whole raw cuts, not ground, and both sit in the calculator as the lower-fat comparison point against the two beef blends.

CutCalories (kcal)Fat (g)Protein (g)Iron (mg)FDC ID
Lamb, domestic, leg, whole, raw23017.217.911.66174311
Goat, raw1092.320.62.83175303

What the numbers say

  • fat is where the four separate most clearly — from 2.3 g in goat to 17.2 g in the whole lamb leg, a wider spread than the two beef blends show against each other
  • iron does not track fat the same way: goat, the leanest of the four, also carries the most iron per 100 g; lamb leg, the fattiest, carries the least
  • which cut a recipe should reach for is still the anatomy question covered above, not a question the nutrition table settles

Which cuts suit which treatment is a question with an anatomical answer, even when tradition gets there first.

— Idris Fenn

Reading

Session 4’s tutorial exercises the macro calculator on two ground-beef blends, lamb leg and goat. Next: week 5 moves from anatomy to what processing does to aromatics.