Pages

Wednesday, October 30, 2013

Leaf scorch caused by pecan anthracnose

   This fall you might have noticed pecan leaves with brown, crispy margins (photo at right). This past summer's rainy weather helped to promote the spread of this disease once known as "fungal leaf scorch". Now, during the month of October, this disease has caused the early defoliation of some pecan trees.
    Leaf scorch is actually just one symptoms of a late-season disease called anthracnose which attacks both pecan leaves and nuts. Pecan anthracnose is caused by the fungus, Glomerella cingulata, a common plant pathogen that causes diseases in many fruit and vegetable crops.
   On pecan leaves, anthracnose first appears as brown, irregularly shaped lesions along the edges of a leaflet. These lesions can spread rapidly over the entire leaflet ultimately causing early leaf fall. The advancing margin of the infection forms a distinctive dark brown line that separates healthy tissues from disease killed tissue (photo above).      


    Anthracnose infection of the nut can start as small sunken lesions on the shuck but the disease can spread to cover a large part if not the entire shuck (photo at left). Infections that cover the entire shuck by early August (water stage) can cause smaller nut size and prevent normal shuck split. Anthracnose infections that occur late in season seem to have little impact on the nuts other than to advance the shuck opening process.


   To determine the influence of anthracnose on nut quality, I collected nuts with infected shucks, peeled them out, then cut them in half to check kernel fill. The nuts that appear in the photo at right represent a range of anthracnose disease severity--from less disease on the left to the most severe on the right. The nut that came out the the shuck appears directly below the nut-in-shuck photo. Below each nut is that same nut cut in cross-section.
    The first thing I notice in the photo is that all 4 nuts are well filled with kernel. Looks like this year's anthracnose infection on nuts will have little effect on kernel quality. However, the nut on the far right was smaller than the others and the shuck did not peel off very easily. In this case, anthracnose probably got an earlier start on colonizing this particular nut.
    Even though the kernel inside the nut on the far right looks good, this nut would probably not survive the nut cleaning process. Judging from the force I needed to apply to remove the shuck, this nut would probably be discarded off the cleaning table as a stick-tight. If anthracnose causes enough stick-tights, this disease can lead to significant yield losses.
    Anthracnose can be controlled with fungicides and is largely suppressed when we spray fungicides to control pecan scab. However, this year's unusual weather patterns hit just right to promote an outbreak of anthracnose after we finished making fungicide applications aimed at controlling scab.  

Monday, October 28, 2013

Nut development: 28 Oct. 2013

   Each time I checked the development of Maramec nuts this summer, I wondered if they would have enough time to split shuck before the first killing freeze. Well, I got my answer today. I found the shucks split open on our scab-covered Maramec nuts (photo at right).
   Splitting shuck before the first freeze is important but  remember, Maramec was having a hard time filling out its kernel ever since mid-September.  So I collected a few Maramec nuts to check kernel quality. Since scab was also an issue with Maramec this year, I harvested nuts with varying amount of scab infection on the shucks.
   
     The photo above shows four Maramec nuts arranged by severity of scab infection. Below each nut is a photo of that same nut in cross section. Nut "A" had the least amount of scab and achieved normal size for a Maramec. A look inside nut "A" reveals less than perfect kernel fill with air pockets both within the kernel and near the inner-shell partition. The poor kernel fill observed inside nut "A" was a result of this cultivar running out of time and heat to fill the seed. In this case, scab infection was not severe enough to negatively affect kernel fill.
     Nut "B" is about the same size as nut "A" but total disease coverage on the shuck of nut "B" has caused even poorer kernel fill.  The earlier in the year scab covers the shuck, the greater effect the disease has on nut size.  Nuts "C" and "D" are examples of the how nut size can be affected by scab. Early scab infection had the greatest impact on nut "D" decreasing size and severely limiting kernel fill.

Thursday, October 24, 2013

Pecans ripe but not ready

Gardner, 9 Oct. 2013
   Two weeks ago I posted this photo of the 'Gardner" pecan cultivar (photo at right). By October 9th the shuck of Gardner had split and this pecan fruit was botanically ripe. However, if you were to remove the nut from the husk at this time and immediately crack open the shell you would find a white, rubbery kernel inside. And the nut meat would taste nothing like the sweet, oily kernel you normally associate with fresh pecans.

Gardner, 23 Oct. 2013
   Two weeks later, the shucks on Gardner look about the same as they did before. The shucks might be slightly more dry along the split-open sutures but the hull is still green and covering the nut (photo at left). I pulled a nut from the husk and shelled it out. The kernel was not at all dry and still hadn't developed its normal golden color (photo above). Two weeks after "ripening" this Gardner was still not ready to eat.
    The bottom line is that the green shuck traps moisture in the nut, preventing it from fully curing. Under natural conditions, our trees will experience a hard freeze, killing the green tissues in the shuck and allowing both shuck and nut to fully dry before harvest.
     For those not patient enough to wait for a killing freeze, the nuts can be physically removed from the shuck (by hand or with a mechanical deshucker). Once the nuts are removed from green shucks, they are still full of moisture and will need to be dried.  I've see all kinds of methods to air dry pecans. Nuts can be placed in mesh bags and hung from the rafters in the garage. Pecan can also be spread onto drying racks built from 2x4s and 1/2 inch hardware cloth. If you use drying racks, make sure to allow for free air movement from both above the rack and below the hardware cloth bottom. Don't use heat to dry pecans but a simple box fan is useful for keeping air moving around the nuts.
     On a larger scale, I've seen growers use a "peanut" wagon to dry pecans. A peanut wagon is basically a large drying rack on four wheels. A grain drying fan (no heat!) is attached to the wagon to keep air moving around the nut crop and promote drying.
     Personally, I prefer to allow nature to dry my pecan crop. Natural drying is a heck of a lot less work and doesn't require the purchase of additional equipment or electricity  

Wednesday, October 23, 2013

Heavy pecan crop + poor limb structure = limb breakage


    The photo at right is a great example of what can happen if a young tree doesn't receive proper training early in its life. This 'Gardner' pecan tree is only 11 years old  but was loaded down with nuts. Under the strain of a heavy crop, one large limb simply couldn't support the additional weight and broke from the tree. This tree has now lost nearly one third of its nut bearing area and it will take several years to fill out the canopy again.
     A close-up photo of the limb's breaking point reveals exactly why the limb couldn't support a heavy crop load (at left). You can clearly see evidence of a bark inclusion, a key characteristic of a narrow crotch angle. This type of branch attachment is structurally weak and prone to breakage under the stresses of wind, ice, or heavy crop load. In a series of blog posts on training young trees, I give tips on how to avoid growing trees with narrow branch angles by a process I call directive pruning and some simple pruning guidelines I call the 2-foot rule. The training young trees series starts here.

Monday, October 21, 2013

Pecan cultivars ripening by October 21

Greenriver, 21 Oct. 2013
   This morning, I checked for shuck split on several of our mid-season ripening pecan cultivars. It seems that many of the cultivar we usually see split by the second week of October are about a week late. Today, I photographed Giles, Jayhawk, Greenriver, Oswego, and Chetopa (photos above and below).

Giles, 21 Oct. 2013



Jayhawk, 21 Oct. 2013

Oswego, 21 Oct. 2013


Chetopa, 21 Oct. 2013

Thursday, October 17, 2013

Evaluating pecan cultivars in SE Missouri.

Kanza, 17 Oct. 2013
    I traveled down to the Bootheel of Missouri to check on a cultivar trial we established near New Madrid. Most of the cultivars under test had split shuck by 17 Oct. 2013 with the exception of Oconee.  I peeled out an Oconee nut and it looked to be at least a week away from ripening. This year, Kanza, Lakota, and Gardner were the most impressive in terms of yield. Photos of some of the cultivars we have under test can be found above and below.

Gardner, 17 Oct. 2013
     In addition to collecting nut samples from each cultivar in the test, I did learn that raccoons love the taste of pecans.  It looks like the 'coons devoured every nut from the Oswego, City Park, and about one half of the Pawnee trees. Unlike squirrels that start eating pecans as soon as they enter the dough stage (mid-August), raccoons wait until shuck split to steal nuts. Once the shuck splits open, a raccoon will peel off the upper portion of the shuck, snatch out the nut,  and leave the bottom portion of the shuck as evidence of their thieving ways.

Lakota, 17 Oct. 2013

Surecrop, 17 Oct. 2013

USDA 75-8-5, 17 Oct. 2013

Oconee, 17 Oct. 2013

Wednesday, October 16, 2013

Broken twigs caused by long-horned beetle

   This week, I've been back up in pecan tree canopies collecting nut samples from our breeding plot. When I'm up in the hydraulic lift I get to see things I might not notice from ground level. High-up in one tree, I found a branch that looked like it was cleanly broken in two with the leaves above the break dessicated to a crispy brown (photo at right).

     Moving in for a closer look, I found that the branch had a clean break--almost like it was partially cut by a pruning tool. This type of clean break is the tell-tale sign that the limb had been attacked by a long-horn beetle called the twig girdler. You can find a photo of an adult twig girdler and the details of this beetle's life history in a post I made previously.
     The female beetle lays eggs in the branch above the girdle. These eggs will hatch next spring and larvae will feed on the recently killed twig. If you find neatly broken off limbs on the ground under your pecan trees this harvest season, it is a good idea to pick up those twigs and burn them to destroy next year's crop of twig girdlers.

Monday, October 14, 2013

Nut development: 14 Oct. 2013

Kanza, 14 Oct 2013
   It been a couple of weeks since I posted an update on the development of the three pecan cultivars that I have been following this season: Osage, Kanza, and Maramec. Osage completed nut development back in late September. Since that time, I've been waiting for Kanza to show the first signs of shuck split.  Today, I found a few nuts in every cluster of Kanza nuts with open sutures (photo at right). In fact, if I pressed on shucks that were not obviously split open, I could cause the shuck to pop open along the sutures. This brings up an interesting point about the development of normal shell color as it relates to shuck dehiscence.


 




     The  photo at left shows the normal progression in shell color development. When the shuck first separates from the shell the sutures are still tightly closed. At this point, the shell is mostly white but marked with the characteristic black speckles and streaks. The first indication of shuck split occurs when a small opening develops at the apex of the nut. As fresh air reaches the shell it oxidizes and begin to turn brown in color (2nd and 3rd nuts in the progression pictured above). As the shucks split downward, from apex  to nut base, more air surrounds the shell causing the development of natural shell color.  Finally, when shucks are split wide open and the nut inside becomes fully colored.

Maramec, 14 Oct. 2013
     Maramec has completed kernel fill in spite of the scab infection that covers the shucks (photo at right). It will be interesting to see if this cultivar will shuck split before first frost.
    Kanza seems to be about 10 days later than normal. If Maramec is also 10 days late we'll need freeze-free weather into early November to see shuck split.












        I cut into the shuck of a Maramec nut and found no trace of shuck separation at this time (photo at left). The amount of scab on this nut should not impact normal shuck dehiscence so, if and when, Maramec splits open we should be able to record that date.

Sunday, October 13, 2013

Pecan diversity

Nuts from seedling pecan trees compared to Pawnee (circled)





    Most pecan growers understand genetic variation in pecan simply because they have grafted a dozen or so cultivars into their orchards. We have large pecans and smaller pecans. There are cultivars that bear heavily and those that are shy producers. Some cultivars seem to have a crop every year while other trend towards alternate bearing. Everyone has grafted cultivars they wish they hadn't, and sometimes we wish we had grafted more of a particularly outstanding nut.
    The genetic diversity within pecan is amazingly wide. In our breeding plot, we used Pawnee as one of the primary parents in making controlled crosses. In the photo above, I've arranged some of the seedling nuts we have collected that ripen before or at the same time as Pawnee. The two nuts circled in the center of the photo are Pawnee nuts to give you a reference for comparison.
    There are a couple of seedling nuts that remind of the Pawnee parent, but the vast majority are drastically different in size and shape. Some of these seedlings will have thick shells while others will be thin shelled. But remember, nut size and percent kernel are just two traits that we need to look at in searching for new pecan cultivars. Disease resistance, good tree structure, and high yield are three major traits are also on the top of my wish list. Its no wonder that new, exceptional, pecan cultivars are so hard to come by.

Thursday, October 10, 2013

Hickory shuckworm: The worm in the shuck.

   At this time of year, the shucks on many pecan cultivars are starting to split open. Like every over-anxious nut grower, I can hardly wait  to see whats inside the the shuck so I pull a pecan from the tree and peel off the shuck.
    In pulling off the shuck, many times I find an area of blackened tunnels carved into the shuck (photo at right). These tunnels in the shuck were created by an insect called the hickory shuckworm.
    If you carefully tear away the shuck you can often find the small white caterpillar with a red head that had created the tunnels (photo at left). Even though this insect can tunnel a large portion of the shuck , the early ripening nature of our northern pecan cultivars means kernel filling is largely completed before tunneling becomes extensive. This means late season hickory shuckworm feeding does not pose a serious economic threat in our area.
   If hickory shuckworm causes any damage it is usually superficial discoloration on the outside of the shell (photo at right). However, when pecans are mechanically harvested, the constant tumbling inside the equipment usually rubs off all shell markings including shuckworm discoloration.
   
    Once the shuckworm larva reaches maturity, the caterpillar carves a trap door in the shuck (large black spot on shuck at left). The larva then pupates right under the trap door. Once the pupae matures, an adult hickory shuckworm moth emerges from the pupal case and the moth exits the shuck through the pre-made trap door.
    The adult hickory shuckworm is a small (3/16") dark moth that has an elongated bell shape when at rest (photo at right). Hickory shuckworm moths are largely nocturnal and you rarely see them during daylight hours. I found this moth resting on the shuck just after it had emerged from the shuck.

Wednesday, October 9, 2013

Pecan cultivars that split shucks in early October 2013

Pawnee, 9 Oct. 2013
   After spending several days in the pecan breeding plot, I've finally had the opportunity to photograph more pecan cultivars as they ripen at the Pecan Experiment Field this year. Today I found Pawnee, Gardner, and Faith had split shuck. Two older northern cultivars, Posey and Peruque, had also split open. Notice the heavy scab infection on the Peruque nuts.
      And this is the first year we've had a good look at  a Stark Bros. pecan cultivar called Surecrop. Photo's of these cultivars are at right and below.
Gardner, 9 Oct. 2013

Faith, 9 Oct. 2013

Peruque, 9 Oct. 2013

Posey, 9 Oct. 2013

Surecrop, 9 Oct. 2013

Tuesday, October 8, 2013

Don't judge a pecan by its cover

     Over the past several days, I've been in a hydraulic lift collecting pecans from our pecan breeding plot (photo at right). Moving from tree to tree, it is amazing to witness the genetic diversity of pecan. Fortunately, we used early maturing parents for this project and now, 15 years later, most of the trees in this planting have ripened their nuts by the first week of October (not all though).
    As you might expect, I have found nuts of all sizes and shapes. Some have thin shells. Some will need a sledge hammer to crack them open. In addition, there are huge differences in reaction to leaf and nut diseases. After looking at so many seedling pecan trees with such diverse sets of cultivar traits, I now have a greater appreciation of how rare it is to find a truly exception new pecan.

    Today, I collected nuts from five different trees just to give you an idea of the type of variation I am seeing in our breeding plot (photo at left). All of these nuts had achieved shuck split with nut "E" probably splitting the earliest. I choose these five pecans because they were roughly the same size in the shuck. When pulling these nuts from the tree, these five nuts appeared to be fairly large. But looks can be deceiving.
   In the photo at right, you can see the same five nuts pulled out of their shucks.  The differences in size and shape are obvious. A big shuck doesn't always produce a big nut (nut C) and surprisingly large nuts can fall from not-so-impressive shucks (nut E).
   The shuck is only the first cover that needs to be peeled back to reveal a pecan's true character. The second cover is the shell. In the photo at left, I've arranged the same five nuts in the same order, but this time, each nut is cut in cross section. Note the differences in shell thickness.  Nut "A" has the thinnest shell, while nuts "D" and "C' have heavy shells. Now look at the packing material as it dips down into the dorsal groves of each kernel half. Long narrow dorsal groves means that the packing material might get stuck in the kernel during the cracking process. Nuts "C" and "E" might have a problem in this regard.
    Overall nut quality for 2013 looks great. You can look forward to seeing all the nuts we collect this year from the pecan breeding plot on display at the nut exhibits held in conjunction with the KS, MO and IL nut growers annual meetings early next year.

Friday, October 4, 2013

Time for Fall fertilization


    With an 85% chance of rain tonight, today was a good day to spread some fertilizer in our pecan orchards (photo above).  An early October application of nitrogen  along with the regular springtime fertilizer application has been our standard practice for over 15 years now and has led  to increased yields and a reduction in alternate bearing. Today we spread 100 lbs. of urea fertilizer/acre. That works out to 46 lbs of actual nitrogen per acre.  The costs associated with this fertilizer application were $411/ton for the urea and $8/ton for the rent of the spreader. Doing the math, we invested $20.95/acre today in building the productive capacity of our pecan trees.
     Next March, we'll be back with the fertilizer buggy to make a springtime application of both nitrogen and potassium. 

Wednesday, October 2, 2013

A story about resistance to pecan scab

USDA 75-8-5
    I was out looking at the pecans in our cultivar trials and came across something I thought was pretty interesting. We have two USDA clones under trial that originate from the same cross; Osage x Creek. The clones are numbered 75-8-5 and 75-8-9. The 75 represents the year (1975) the cross was made, the 8 represent the parents (in this case, Osage x Creek), and the final number is the number given to the seedling tree within all seedlings of that cross.

USDA 75-8-9
       USDA 75-8-5 is pictured above while USDA 75-8-9 is pictured at left. Wow, the difference in scab susceptibility hit me like a brick. Seeing this obvious difference between siblings in the amount of scab found on the nuts made me starting thinking about the inheritance of scab resistance. This took me on a quest to look up the ancestry of these two USDA clones.



       Above is a flow chart of the ancestry of USDA 75-8-5 and USDA 75-8-9. The parents of these clones were the same: Scab resistant, Osage and scab susceptible, Creek. Going back another generation I found that the scab resistance of Osage originated from its Major parent. Looking at the ancestry of Creek, I found scab susceptibility in all cultivars that make up the Creek side of the family tree.
    I don't think that scab resistance is a single gene inheritable trait because we see such a wide range of reactions to the scab fungus. However, a quick look at the family tree would lead me to expect that only a portion of Osage x Creek crosses would demonstrate some level of resistance to scab.
    Even though 75-8-5 looked scab free in our trials in 2013, this USDA clone has demonstrated only mediocre scab resistance under conditions of  high humidity in southern Georgia; a location of much higher scab pressure than SE Kansas (my location).

Tuesday, October 1, 2013

Nut development: 1 Oct 2013

Osage
   Again this week I looked at the development of three pecan cultivars: Osage, an early-ripening cultivar; Kanza, a mid-season pecan; and Maramec, a late-ripening cultivar. As noted in yesterday's post on cultivars ripening since last week, Osage has completed the nut development process for the year by splitting open its shuck (photo at right).
Kanza
    Kanza has filled it kernel and I found the very first indications that the shuck of this cultivar was starting to separate from the shell. In the photo at left, the red arrow points to the area where the shuck has started to separate from the shell. At this point, all you can see is a hairline crack between shuck and shell. Remember, that during shuck dehiscence, the process always begins at the nut's apex and works downward towards the base. The green arrow points to another indication that shuck separation has begun. The light tan spots near the tip of the nut are the very beginnings of black streaks normally found on the outside of the shell at harvest.
Maramec

     I think that Maramec has largely completed kernel fill and you can see that this year's nut quality will be fair to poor (photo at right). The kernel has failed to press all the internal packing material up firmly against the inside of the shell and there are still air spaces near the kernel. You might also note that this Maramec nut is much smaller than normal which can be largely attributed to scab infection. At this point in time, Maramec is not showing any signs of shuck dehiscence. Will we harvest Maramec this fall? It will probably depend on when we get our first hard freeze this fall and if the scab infection will prevent shuck split.