Investigating the Effects of Increasing Copper Sulphate Solution Concentrations on the Germination of Cress Seeds

---

I will then prepare stands for the seeds to germinate on.

Each stand will be prepared as follows:

---

---

Null Hypothesis

There is no significant difference between the number of seeds which germinate in the differing copper sulphate concentrations 

---

Raw Data

---

Manipulated Data

---

I will now use the values and the quantities worked out for my manipulated data to work out if the means of the samples are significantly different. As I have worked out the standard errors and the means, I will first use the 95% confidence standard error test. I am using this because it is used to work out if means of different samples are significantly different in normal distribution values.

---

Analysis of Standard Error Test

This table below interprets the log of concentration graph to show where there were and were not significant differences between the mean number of seeds germinated at each differing cupper sulphate concentration:

Having done the 95% confidence standard error test I would reject my null hypothesis. This is because there were significant differences between the means of the number of seeds germinated per concentration at the 5% level of probability. From the standard error test, it shows that there was a 95% chance that the means were significantly different between the concentrations marked above in the table. But because the standard error test is usually suitable for sizes of samples which are at least 30, and unfortunately, due to limitations we could only do 8 repeat samples, this may not be entirely reliable. So I will also do the t-test, which is another test for finding out if there is a significant difference between two means in normally distributed data, but for samples smaller than 25.

---

I will now use the values and the quantities worked out for my manipulated data to work out if the means of the samples are significantly different. As I have worked out the standard errors and the means, I will first use the 95% confidence standard error test. I am using this because it is used to work out if means of different samples are significantly different in normal distribution values.

---

T-Test

For the t-test I will compare closely the results from the 60mg/l, 0.06mg/l and 0mg/l tests. This is because the 0.06mg/l is the natural level for cress seeds to germinate in when in their natural environment in the soil. The 60mg/l is the strongest solution of copper sulphate I used, and the 0mg/l is the pure water solution, which shows germination with no excess copper sulphate.

---

Table of t distribution

---

Analysis of t-test results

The 0mg/l and 0.06mg/l test:

The t value was 1.28 and the degree of freedom was 14. This meant that I had to find the probability using the following column of the t distribution table:

This shows that p>0.10 and so the results of these two concentrations are not significant, so the null hypothesis is accepted here as the mean number of seeds germinated between the concentrations is not significantly different.

The 60mg/l and 0.06mg/l test:

The t value was 16.9 and the degree of freedom was 14. This meant that I had to find the probability using the following column of the t distribution table:

This shows that the p<0.001. This means that the results are very highly significant and that I can reject my null hypothesis and be almost certain that the mean number of seeds germinated in concentrations 60mg/l and 0.06mg/l are significantly different.

So I have rejected my null hypothesis.

---

From using both the 95% confidence standard error test and the t-test, I now reject my null hypothesis. This is because the tests show that the mean number of seeds germinated at differing concentrations of copper sulphate was significantly different.

---

From looking at the manipulated and the raw data, I can see that, from moving down the strength of copper sulphate concentrations, (60mg/l-0mg/l) the mean number of seeds germinated increased with each new concentration to a point, 0.006mg/l with a mean of 14.6, and then as the concentrations of copper sulphate in the solutions get smaller, then the mean number of seeds germinated decreases for each new concentration to 0mg/l and a mean of 11.6. Because the number of samples for each concentration was the same, then this increase and decrease of mean coincided with the increase and decrease of the total number of seeds germinated per concentration over the 8 samples.

At first look, there does not appear to be a trend or pattern in the standard deviations for the 7 concentrations of copper sulphate. But if the 2 concentrations with the highest mean and total numbers of seeds germinating are removed (0.06mg/l, mean 13.3 and standard deviation 1.8 and 0.006mg/l mean 14.6 and standard deviation 3.0) then the standard deviations follow the same trend as the means, with an increase to a point and then a decrease again. This is relevant because the 2 concentrations removed are the 2 closest to the normal value of copper sulphate in soil for cress seed germination to occur. They have lower standard deviations than the other closer concentration strengths. The standard errors follow the exact same trend as the standard deviations.

When looking at the standard error 95% confidence test results and particularly the log of concentration graph showing the means with standard error, I can see clearly the increase and decrease in the means as described above. It is also clear from looking at the results of the standard error test, that all the concentrations had significantly different mean number of seeds germinated than the 60mg/l copper sulphate concentration. But the 2 concentrations with the low standard deviations and standard errors, which were closest to the natural copper sulphate levels, also had significantly different means from the 6mg/l tests.

As I only needed to do the t-test with 3 of the concentrations of the investigation, then no real patterns appeared. Although the t-test did support the fact that I rejected my null hypothesis because it showed a significant difference between the mean number of seeds germinated in some of the concentrations.

---

From the results it can be seen that the favoured concentrations of the cress seeds for germination are 0.06mg/l and 0.006mg/l of copper sulphate solution. The least favoured concentrations are the strongest concentrations of copper sulphate (6mg/l and 60mg/l), although the weakest solutions, such as the pure water, also did not favour cress seed germination as much as the 0.06mg/l and the 0.006mg/l concentrations.

The cress seeds did not germinate well in the strongest concentrations of copper sulphate, because germination is initiated by the seed taking up water rapidly, which causes the seed to swell and thus rupture the testa. The water will then hydrolyse insoluble storage material to soluble substances which can be transported. These are transported to the point at which the embryo is growing. This was prevented from happening in the strongest solutions.

The uptake of water depends on osmosis, which relates to the diffusion of water across a partially permeable membrane from a less negative water potential to a more negative water potential. Seeds normally have a low water potential, it is normally very negative, but require a water potential of around -2Mpa for germination to start. Because they have such a low water potential they normally take in water quite easily. But in the high concentrations of copper sulphate, they difference in water potentials between the solution and the seed is not great, so they do not take in enough water for the testa to rupture and for the hydrolysis of the insoluble storage material or activation of the enzymes for food mobilisation.

The germination also depends on the activation stage which is the mobilisation of foods such as starch proteins and fats. This occurs through enzymes (amylase, maltase, peptidases, lipase) changing the substances. The starch is broken down to maltose, which in turn is broken to glucose and transported as sucrose for cellulose or used for energy. The proteins in the seed are converted to polypeptides and then broken to amino acids, which are used for structural or enzyme proteins. The fats are converted to glycerol which is converted to transportable sugars or broken to fatty acids for energy or for transport. If these enzymes are not mobilised by the initiation of germination then the seed will die, and no radicle will appear.

Also in these high concentrations of copper sulphate, the seed will take in copper sulphate through diffusion down the concentration gradient. Copper acts as an enzyme inhibitor in high concentrations and it is very likely that it prevents the breakdown of stored food products in the seed and also inhibits the formation of chlorophyll. This results in inhibited germination and radicle growth. The copper levels will change the pH of the solution in the seeds, which will move the pH away form the optimum level of the enzymes needed in the food mobilisation. Because the seed has lost it’s carefully maintained optimum levels for the enzyme activity, then the enzymes will be denatured. This is when the pH change causes the tertiary structure of the enzyme to lose its shape, with the ionic bonds breaking up. When the active site loses its shape from this and from a loss of charge in the amino acids, the enzyme is said to be denatured and no longer functions, or forms enzyme substrate complexes.

The cress seeds had the highest mean number of seeds germinating at the concentrations 0.06mg/l and 0.006mg/l of copper sulphate. I researched the optimum conditions for cress seed germination, and found that the natural level of copper in the soil for cress seed germination is 0.064. The two concentrations with highest means are those closest to this value. These concentrations had the most seeds germinating because they had the right concentration to get enough copper through diffusion for chlorophyll formation and for proper enzyme activity. And also these concentrations were low enough to not inhibit enzyme activity or effect osmosis and stop water entering the cell. The concentration 0.06mg/l of copper sulphate, which was closest to the natural level, had the lowest standard deviation other than the 60mg/l of copper sulphate. This shows that in each batch almost the same number of seeds germinated, showing that this concentration was consistently right for seed germination. This emphasizes how the 0.06mg/l was perfect for germination.

The lowest concentrations would not have had as many seeds germinating as the higher concentrations because there would not be enough copper taken in by diffusion to start chlorophyll formation and help with enzyme activites.

---

Anomalies and Improvements

In the raw data, some of the results stand out. These were in the 6mg/l, 0.6 mg/l, and 0mg/l copper sulphate concentrations, and one test had other problems:

These anomalies in the results could be because of a number of reasons.

Although I planned my investigation out carefully, there were some things that I was unable to do as I had planned, and some things which I had planned which should have been done better and there are a few things which I would change if I was going to do the investigation over again.

Although I had planned to repeat each copper sulphate concentration 24 times, due to limitations I could not do this. When I planned the experiment, I planned it so that I would be able to carry out the investigation to achieve the most accurate and statistically valid results, so I planned to do 24 repeats. But as the school could not provide 168 pots or pipette dishes, I could only do 8 repeats of the differing concentrations. I could possibly have done the test 3 times over, but then it would have been three different investigations because it would have meant that the tests were done in different conditions for example different temperatures. Also the solutions would not all have been made together, so there could have been some apparently minor differences in the concentrations, which would have had a big effect on the results of the test. So I would not have been able to use all of the results together. This could have affected the accuracy of the results, and of the statistical tests, because there were not many samples to use, particularly to check the significance of the results using the 95% confidence standard error tests. Because I was not doing as many repeats as I had planned, I also had to change the number of seeds which I was going to put in each pot for each concentration form 15 to 30, so that the test was still statistically valid.

Limitations on time, and availability of usage of the classroom lab also meant that I could not go to the lab when I planned to, to put solutions back into the pots so that they did not dry out. So I decided that I would not put any more solution on any of the pots at all after the investigation had started. This led to the sample number 5 in the 0mg/l copper sulphate concentration drying out. This would have meant that the seeds in that particular pot did not have water available to all of them for the whole time, so the results here were questionable. This is supported by the fact that only 5 seeds germinated in this pot, whereas the average for the 0mg/l concentration was 11.6. But it also seems strange that this was the only pot which did dry out in the whole experiment, this may have been because I was at fault and did not put the full amount of solution into the pot at the start, so it would have been quicker to dry out.

This may also have dried out because I was not able to have access to clean needles to pierce the cling film for air to circulate into the seeds, so I had to use a stirring rod to pierce it. This meant that the wholes were very big, and so evaporation could have occurred. But this should not really have affected the investigation, because all of the pots had cling film pierced with the same size stirring rod, the same number of times.

Another thing I had to change from the plan which would not have affected the investigation because it affected all the seed samples the same was that I had to use smaller pots than I wanted; they were only 15cm in diameter. Also, I had to use syringes to measure out each solution because there were not enough micropipettes for the class to all use one. This could not have affected the investigation because it meant all solutions were being measured the same way, but some may have not been measured very accurately, this may be why I might have measured the solution for sample number 5 in the 0mg/l copper sulphate concentration wrongly.

Although the results for the samples 6 and 7 in copper sulphate concentration 0.06mg/l do not appear to be very wrong, because 10 and 11 seeds germinated in each and the mean number was 13.3, the results were affected by the fact that when I put the seeds onto the filter paper in the solution, some seeds moved about in the solution away from where they were supposed to be and some were very close to each other in the experiment. This would have meant that they were affected by intraspecific competition, as the seeds would have competed for the available water, meaning some might not have got the needed amounts of water for germination to be initiated. If the pots were 20cm in diameter instead of 15cm, then the water would have been shallower and more spread out, meaning the seeds would not have floated at the start.

Something else which may have affected the investigation would have been that although I said I would put up glass screens in front of the pots to stop wind or drafts affecting the temperature and increasing the rate of evaporation into the air, there were no glass screens available for me to use. Therefore, in their place, I wrapped large amounts of cling film around the area with the pots to stop this, and put holes in this facing away from doors for air to circulate, it seemed largely ineffective and fell down quite a bit, not really protecting the seeds in the pots at all. This could have meant that drafts may have increased the rate of evaporation in the pots. But all of the seeds in all of the pots were again under the same conditions stopping the test becoming unfair and keeping it statistically viable. Also I kept a thermometer checking the temperature of the area with the seeds at all times, this was linked to a computer, which drew a graph of the temperature, showing any fluctuations over the time of the investigation. This graph is included over the page. It shows there was a very small 2 degree variation in the temperature and this too would have been experienced by all of the pots together. SO none would have been in different conditions to the other plant pots in the investigation Therefore this would not really have had that big an effect on changing the results of the tests.

If I were to do the investigation again there are some things which I would change, these were mentioned above in the anomalies and the sources of error which occurred with the procedure and the apparatus which I used.

I would do the investigation with more repeats of the samples as I had planned, so that the results would be more statistically viable. I would do this by doing the investigation at a time when I would be able to use all of the pots which I needed, this would mean that I would be able to use them all without disturbing other experiments. I would also make the seed numbers 20, as this would keep the test statistically viable as the 30 did for the 8 samples, but counting out 30 seeds for 168 different pots is unreasonable for the experiment to be done in time.

I would also use an incubator for my tests instead of using glass screens as I planned, or cling film as I had to. This would mean I could keep the temperature constant for the whole experiment and no fluctuations would occur. Therefore, the seeds could be kept at a temperature optimal for the germination of cress seeds.

As I had planned I would pierce the cling film with a needle not a stirring rod so that the wholes were small and all the same size. This would be enough for air to get in and out, and for evaporation to be largely prevented form the solution. I would also use a micropipette to take the measurements because this would make them more accurate and so no miss-calculations would be made. These two precautions in procedure, should prevent any pots drying out like they did in the sample number 5 of the 0mg/l of copper sulphate solution in the investigation. As none of the other pots dried out, I would not have to put more solution on each pot to stop it drying out. This would also mean that I could make a lot less of each solution than I would if I was topping up each pot with more solution. I would only need one quarter of what I needed to plan for.

I would use different pots (20cm in diameter as planned) for my investigation, so that they were big enough to stop the seeds floating in too much solution and to keep them far enough apart, and small enough to stop the solution being to spread out and shallow in places. If this were not possible I would make less solution (6ml for each pot) and still use the pots which were 15cm in diameter.