Contents
Interesting experiments
A result is only as good as the comparison that frames it. These eight are worth knowing for the design.
For each experiment: the question, the comparison, what the result supports, and what it does not. The last part is where exams get lazy and science does not.
Pasteur’s swan-neck flasks
Question. Does broth spoil because life appears in it spontaneously, or because something falls in from the air?
Comparison. Nutrient broth was boiled in flasks whose necks were drawn out and bent down, so air could pass and dust would settle in the bend. Sister flasks were tilted so the dust washed in, or their necks were broken.
Result. Upright swan-neck flasks stayed clear. Broth that met the dust grew cloudy. The liquid was capable of supporting life the whole time. It did not generate it.
It does not show that every infectious disease is bacterial, and it does not describe viruses. It settles spontaneous generation for microorganisms in that broth.
Mendel’s pea counts
Question. When two true-breeding plants differ in a trait, what happens to the trait in the grandchildren?
Comparison. Mendel crossed lines that bred true for contrasting forms — yellow or green seeds, tall or short stems — and counted every class in the next generations, rather than describing a typical plant.
Result. The first generation was uniform. The second generation recovered the hidden form in a ratio close to 3:1. That fits a factor that is present in two copies, of which only one is expressed, and which separates cleanly into gametes.
It does not show that every trait is a single gene with two forms and complete dominance. Height in humans, for a start, is nothing like seed color in those peas. Linkage, incomplete dominance, and environment are outside the design. The design is still the model of a count that can falsify a ratio.
Griffith’s transformation
Question. Can a heritable trait pass from a dead bacterium to a live one?
Comparison. Frederick Griffith (1928) used two strains of pneumococcus. Smooth (S) cells kill mice; rough (R) cells do not. Heat-killed S cells alone do not kill. Heat-killed S cells mixed with live R cells do.
Result. Live bacteria recovered from those mice were smooth, and their descendants stayed smooth. Something from the dead cells — he called it a transforming principle — had changed the live ones heritably.
It does not identify the molecule. Protein, polysaccharide, and DNA were all still candidates. That identification is Avery, MacLeod, and McCarty, and then Hershey and Chase.
Avery, MacLeod, and McCarty
Question. Which substance in the extract is the transforming principle?
Comparison. In 1944 they purified the transforming material and treated it with enzymes. Proteases and ribonuclease left the transforming activity standing. Deoxyribonuclease destroyed it. The preparation was rich in DNA and poor in protein.
Result. DNA was sufficient for transformation, as far as their chemistry could tell.
Skeptics argued that a trace of protein could still be the real agent. The chemical standards of the time could not drive that objection to zero. Hershey and Chase attacked the same doubt with a different label.
Hershey and Chase
Question. When a bacteriophage infects a cell, does the genetic material that enters come from the phage’s protein coat or from its DNA?
Comparison. Phages were grown with radioactive sulfur, which proteins contain and DNA does not, or radioactive phosphorus, which DNA contains and proteins do not. Each kind was allowed to attach to bacteria. A blender sheared off empty coats. A centrifuge separated bacterial cells from the debris.
Result. Phosphorus entered the cells and appeared in the next generation of phage. Sulfur stayed outside. The coat is a syringe. The DNA is the inheritance.
It is about this phage, T2. It does not by itself prove the structure of DNA, and it does not say that all viruses use DNA. Some use RNA. The blender is famous because it is concrete, not because violence is a method.
Meselson and Stahl
Question. When DNA replicates, what happens to the two original strands?
Comparison. Escherichia coli was grown for many generations in heavy nitrogen, so its DNA was dense, then switched to light nitrogen. DNA was extracted generation by generation and spun in a cesium-chloride gradient until it floated at its own density. The three models make different predictions. Conservative replication keeps a fully heavy helix and makes a fully light one. Semi-conservative replication makes hybrid helices after one generation. Dispersive replication makes helices that are mixed within each strand and would not resolve into clean light and hybrid bands later.
Result. After one generation, a single band of intermediate density. After another, intermediate DNA and light DNA. Semi-conservative replication fits. Each daughter helix keeps one parental strand and builds one new strand.
It does not reveal the enzymes. Helicase, primase, and polymerase are later biochemistry. The experiment settles the pattern of the strands.
Luria and Delbrück
Question. Do bacterial mutations to virus resistance appear because the virus provoked them, or were the mutants already there?
Comparison. In 1943 Salvador Luria and Max Delbrück started many small cultures from a few non-resistant cells, grew them up separately, and then plated each culture with phage. If resistance were a response to the phage, cultures of equal size should produce similar numbers of resistant colonies. If a mutation can happen early in one culture and be copied to thousands of descendants, the counts should fluctuate wildly.
Result. They fluctuated. A few cultures had jackpots of resistant cells. Mutation is blind to the later selection. Selection reveals it.
It does not show that directed mutation is impossible in every biology. It shows that the resistance they measured is not explained by instruction from the phage. The fluctuation test is a way of thinking about variance, and it is worth more than the historical anecdote.
Miller and Urey
Question. Can organic building blocks form from simple gases and an energy source, with no cell present?
Comparison. In 1953 Stanley Miller and Harold Urey sealed water with a reducing mixture — methane, ammonia, hydrogen — and passed sparks through it for a period of days. A control without the spark, and the trap that collected products, let them ask whether amino acids were made or merely contaminated.
Result. Amino acids appeared, including several used in proteins.
It does not reconstruct the actual early atmosphere, which current geochemistry treats as less reducing than the 1953 mix. It does not make a cell. It shows that some of life’s small molecules are chemically cheap under the right conditions. Later experiments have varied the gases and the energy source. The origin of a replicating system remains a harder, open problem.