Someone asked me the other day how our ancestors made the transition from having 24 pairs of chromosomes to 23 pairs of chromosomes. I didn't have a great answer then. What I presume would have happened is this sequence of events: One individual in the population has a fusion between two chromosomes (just one of the two pairs probably) in a germ line cell. This single mutation then spreads through the entire population. Either by drift, or by selection, if it was favored. I would guess that while this was happening, the progenitor chromosomes 2p and 2q, were probably present in most individuals initially as well. My guess is that the fused chromosome 2 rose in frequency in the population to fairly high levels, and that 2p and 2q dropped out of the population after that.
Note that chromosome fusion is not unprecedented in other species. The poster child of chromosomal fusion is the muntjac, particularly the Indian Muntjac (Muntiacus muntjak). The female of this species has the lowest chromosome number of any mammal (6) while the male has 7. This extreme reduction in chromosome numbers has resulted from several serial fusions of chromosomes. Take a look at this phylogeny by Wang and Lan (2000), which shows the diploid number of several Muntjac species, going from the probable ancestral state of 2n = 46 to the current number of 2n = 6 in female Indian Muntjacs:
Amazingly, Indian Muntjacs can produce viable hybrids with Chinese Muntjacs, which have the ancestral 46 chromosomes, and that partial spermatogenesis followed, indicating that they may be partially fertile. In my brief foray into the internet, I was unable to find anyone who had posited why these species have had such dramatic reductions in chromosome number.
Besides these mammalian representatives, even weirder games with chromosomes are played by certain Middle Eastern Toads, which turn out to be triploid rather than diploid. Triploidy isn't actually that uncommon in nature. In fact, some fisheries will induce triploidy in trout by using high pressure to produce sterile fish that they can then stock in waterways without worrying about them spreading. A common means by which triploid species are made is by hybridization. Improper meiosis results in a normal haploid gamete meeting an abnormal diploid gamete. The resulting individual may grow up fine, but it is mostly out of luck when it comes to sexual reproduction, because there is difficulty in the chromosomes lining up and dividing properly. Yet all is not lost. These species could simply just clone themselves by a process called parthenogenesis. A particular remarkable example of this is found in Ambystoma salamanders of the Jefferson-Laterale complex. Oftentimes these species are "sexual parasites" of their parental species. They still require mating as a stimulus, but do not incorporate their partners sperm into their offspring. Sometimes they sexually reproduce with a normal individual, or another hybrid, but the result is an even greater increase in ploidy. Consequently, you can find triploid, pentaploid, heptaploid, and nonaploid salamanders.
But what about the toads? It turns out they've gotten over the problem sexually reproducing. These triploid toads are sexually reproduce and have both males and females. Essentially, one of the chromosomes becomes like a mitochondria, it's only inherited on the maternal line, while the other pair are passed on normally. Therefore the egg is 2n, while the sperm is only n. Here's a recent story on this system.
It's true that a lot of diseases can be caused by chromosomal mutations such as this, but occasionally it appears it isn't that big a deal. It's also certainly possible that some of these mutations were at some point adaptive, and that facilitated their spread through the population. Pretty amazing stuff really.