Regarding the first question:
[a] is incorrect because every point on the CML is a combination of the rf-asset and the market portfolio. The only thing that distinguishes one point from another on the CML is the amount of capital a particular investor allocates to the rf-asset and the market portfolio. Hence all portfolios are related on the CML because they contain the same market portfolio.
[b] is incorrect for a similiar reason. Each point on the CML contains some proportion of capital invested in the market portfolio. You can invest all your capital in the rf-asset, or the risky portfolio. Hence these portfolios can be thought of as distinct from each other. But most of the portfolios on the CML are some combination of these two extremes. Hence we cannot say that ALL portfolios are distinct from one another.
[c] is correct by process of elimination. Another way to see that all portfolios on the CML are positively correlated comes from the fact that all points on the CML are combinations of a rf-asset and the market portfolio and that only the market portfolio has an impact on the risk of the total portfolio (rf-asset has zero risk)
With the existence of the risk free asset the mean and the variance for a portfolio consisting of the risk free asset and the portfolio M (see figure) will be:
Rp = aE[rm] + (1 - a)rf
Hence the total portfolio return is linearly related to the weight in the market portfolio. This can be seen simply from the fact that the CML is a straight line with a constant slope.
Varp = a2VAR[rm]+ (1- a)2VAR[rf]+ 2a(1- a)COV[rm,rf]
= a2VAR[rm]+ (1- a)20+ 2a(1- a)0
= a2VAR[rm]
Hence the variance of a portfolio on the CML depends only on the risk of the market portfolio. As we move from the left to the right on the CML, we are increasing the weight in the market portfolio (a) and hence increasing the Variance of the total portfolio. As we move from the right to the left on the CML, we are decreasing the weight in the risky portfolio and hence decreasing the Variance of the total portfolio.
Theres is a postitive correlation between the weight in market portfolio (a) and the return on the total portfolio
There is a positive correlation between the weight in market portfolio (a) and the variance of the total potfolio.
Regarding question 2:
http://books.google.co.uk/books?id=pZfeJ75ZRJ0C&pg=PA291&lpg=PA291&dq=why+does+market+portfolio+include+all+risky+assets?&source=bl&ots=5eaQ0RAe0Y&sig=e5QtSGYPNrT2sNvipX2E2kK-DHc&hl=en&sa=X&ei=gt6QUO2tMsGi0QXFnoGwBw&ved=0CDMQ6AEwAw#v=onepage&q=why%20does%20market%20portfolio%20include%20all%20risky%20assets%3F&f=false
There is a small snippet for market portfolios here.
hope this makes senseā¦L1 was some time ago