1The idea in one paragraph

A fertilised egg divides into a ball of identical, unspecialised cells. Signals that vary in concentration across the early embryo tell each cell where it is, and that position decides which genes it switches on, so cells set off down different paths and differentiate into specialised types. Some cells stay unspecialised for life: these stem cells can keep dividing and can still differentiate, and they sit in protected niches such as bone marrow and hair follicles, where they replace worn-out cells. Stem cells lose options as development proceeds, from totipotent to pluripotent to multipotent. Specialised cells range in size from a red blood cell a few micrometres across to a neuron a metre long, but no cell can grow large in every direction, because its volume grows faster than its surface area, and the surface is where all exchange happens. At HL, cells beat that limit by flattening, folding their membranes, or growing long and thin, and the lungs, muscles and gametes show specialisation at its most extreme.

2From one cell to many kinds: differentiation

After fertilisation the zygote divides by mitosis again and again. At first the cells produced are unspecialised: they look alike and could each become many things. Later, cells become different from each other in structure and function. This process is differentiation, and it produces the specialised cells of the adult: muscle, nerve, blood, skin and the rest.

Every one of these cells has the same genes, because they all came from the zygote by mitosis. So differentiation is not about which genes a cell has; it is about which genes it expresses. A muscle cell switches on the genes for contractile proteins; a red blood cell precursor switches on the genes for haemoglobin.

How does a cell in an early embryo know which set to switch on? A major part of the answer is gradients. Certain signalling molecules (often called morphogens) are released from one region of the embryo and spread outwards, so their concentration falls with distance from the source. Each cell detects the concentration where it sits, and different concentrations switch on different genes. Figure 1 shows the principle.

Cells near the source receive a high concentration and express one set of genes; cells further away, at a medium concentration, express a second set; cells far away express a third. So a single gradient divides identical cells into distinct kinds according to their position. In the fruit fly embryo, for example, a gradient of one protein from the front end of the egg sets out which end becomes the head. Gradients along two or three directions at once give each cell a unique address, and the body plan is laid down from those addresses.

3Stem cells and their two properties

A stem cell is an unspecialised cell with two properties, and the guide limits you to exactly these two:

1. It can divide endlessly. It keeps its ability to divide by mitosis, producing more stem cells, so the supply never runs out (self-renewal).
2. It can differentiate along different pathways. Its daughter cells can become one of several specialised cell types, depending on the signals they receive.

Stem cells matter in two ways: in the embryo they build every tissue, and in the adult they replace cells that wear out or are damaged. Your body replaces millions of red blood cells every second and renews the lining of your gut every few days; both depend on stem cells.

4Stem cell niches in adult humans

Adult stem cells are not scattered at random. They sit in particular places in tissues called stem cell niches, where the neighbouring cells and the chemical signals around them control what the stem cells do. A niche can do two opposite things:

• maintain the stem cells: keep them unspecialised and dividing only slowly, so the reserve is not used up;
• promote proliferation and differentiation: when more cells are needed, signals in the niche make the stem cells divide faster and their daughters differentiate.

The guide asks for two example locations. Figure 2 shows both.

Bone marrow. The red marrow inside bones contains haematopoietic stem cells, which produce every type of blood cell: red blood cells, all the types of white blood cell, and the cell fragments called platelets. After blood loss or during an infection, the niche signals the stem cells to increase production. This is why a bone marrow transplant can rebuild the entire blood system of a patient whose own marrow has been destroyed.

Hair follicles. Each hair follicle has a region called the bulge that holds stem cells. At the start of each cycle of hair growth, signals activate them to divide and produce the cells that build a new hair. They also help repair the skin after a wound.

5Totipotent, pluripotent and multipotent

Potency is the range of cell types a stem cell can become. It falls as development proceeds, and there are three levels you must distinguish, shown in Figure 3.

The guide's summary is the one to remember: cells in early-stage animal embryos are totipotent but soon become pluripotent, whereas stem cells in adult tissue, such as bone marrow, are multipotent. Each step closes off some paths. A haematopoietic stem cell can make any blood cell, but it cannot make a neuron.

6Cell size as an aspect of specialisation

Human cells vary enormously in size, and size is part of how a cell is adapted to its job. The guide lists the cells to consider; Figure 4 puts them on one scale. The scale is logarithmic because the range is too wide for any other: each gridline is ten times the one before.

• Red blood cells are small, about 7–8 µm across. Small size lets them squeeze through capillaries only slightly wider than they are, and keeps every molecule of haemoglobin close to the surface.
• White blood cells are a little larger, roughly 7–20 µm depending on type. Many must be big enough to engulf bacteria.
• Sperm are among the smallest cells by volume: a head about 5 µm long, and about 55 µm long including the tail. A small cell is cheap to make in huge numbers and easy to propel.
• Egg cells are the largest cells by volume, about 0.1 mm across, just visible to the naked eye. Their size is a store of cytoplasm for the early embryo.
• Neurons have small cell bodies but can be extremely long: the axon of a motor neuron running from the spinal cord to the foot is about a metre long. Length is the whole point, carrying a signal from one end of the body to the other in a single cell.
• Striated muscle fibres are 10–100 µm wide but can be several centimetres long, so one fibre can shorten along much of the length of a muscle.

Notice the pattern: the cells that are very long are also very thin. Section 7 explains why.

7Surface area-to-volume ratio and the limit on cell size

Everything a cell takes in, oxygen, glucose, water, ions, and everything it gets rid of, carbon dioxide and other waste, crosses its plasma membrane. So:

• the rate of exchange depends on the cell's surface area;
• the need for exchange depends on its volume, because volume is where the metabolism happens.

The problem is that when a cell grows, its volume increases faster than its surface area. The guide's model for seeing this is a set of cubes (nature of science: a simplified model of a complex system). Figure 5 works it through.

For a cube of side L:

Double the side from 1 to 2 and the surface area goes up four times, from 6 to 24, but the volume goes up eight times, from 1 to 8. The ratio halves, from 6 : 1 to 3 : 1. Double again, to 4, and it halves again, to 1.5 : 1. Every time a cell doubles in width, each unit of its volume has half as much membrane to supply it.

So a cell cannot grow indefinitely. Past a certain size, its surface cannot take in oxygen and nutrients, or remove waste, fast enough to meet the demands of its volume, and the cell stops growing or divides. This is one reason why large organisms are made of many small cells rather than a few big ones.

Why cubes are a fair model. Real cells are not cubes. But the scaling works the same way for any shape: if you scale any object up by a factor k, its area increases by k² and its volume by k³, so its SA:V falls by a factor of k. The cube is simpler than a real cell, and that is exactly what makes it useful; it keeps the one relationship that matters and drops everything else. A good answer to an "evaluate the model" question says both: the proportions carry over, but real cells are not uniform cubes, and real membranes are not uniformly permeable.

Sample calculation. A cube-shaped cell of side 20 µm has SA:V = 6 ÷ 20 = 0.3 µm⁻¹. A cell of side 40 µm has SA:V = 6 ÷ 40 = 0.15 µm⁻¹, half as much. Give the ratio as a number with the unit µm⁻¹, or as a ratio such as 0.3 : 1; both are accepted if you are consistent.